zugmann I think sometimes we think what is common knowledge isn't so.
Common knowledge, like common sense, is mostly uncommon, it would seem.
BaltACD I have my doubts that it was ever a 'level' crossing - traditionally railroads have built their lines on a bit of a fill so as to enhance drainage of water away from the right of way. Just like highways are constructed with a crown to to drain water off the traveling surface to the sides of the road.
I have my doubts that it was ever a 'level' crossing - traditionally railroads have built their lines on a bit of a fill so as to enhance drainage of water away from the right of way. Just like highways are constructed with a crown to to drain water off the traveling surface to the sides of the road.
BaltACD The hump over the tracks is painfully obvious to someone that can see and think. Obviously the driver was not able to do either.
It also appears that the construction guys got the rig a good part of the way 'off' as the impact on the far track never involves the container nose, whereas in the first shot the tractor is wholly over the crossing. That at least suggests that with only a couple more minutes the truck could have been pulled the 10 feet or less back, or actually started and driven back, to clear.
tree68 charlie hebdo So does the railroad have any responsibility for the impact the changes it made in its ROW have on the crossing? I would opine that, based on the usual arrangements for crossings - ie, the railroad was there first thus the road is the interloper - that the railroad has no legal responsibility to coordinate their actions with the highway people. On the other hand, being a good neighbor might dictate some sort of cooperation, particularly if there are a number of such crossings when a highway parallels a ROW. I'm going to defend MC - he's got a lot of years in the railroad business, much of it work in relation to incidents such as this. You don't tolerate people who offer opinions, etc, that run counter to your profession - he does that same.
charlie hebdo So does the railroad have any responsibility for the impact the changes it made in its ROW have on the crossing?
I would opine that, based on the usual arrangements for crossings - ie, the railroad was there first thus the road is the interloper - that the railroad has no legal responsibility to coordinate their actions with the highway people.
On the other hand, being a good neighbor might dictate some sort of cooperation, particularly if there are a number of such crossings when a highway parallels a ROW.
I'm going to defend MC - he's got a lot of years in the railroad business, much of it work in relation to incidents such as this. You don't tolerate people who offer opinions, etc, that run counter to your profession - he does that same.
I simply raised question. Someone with his expertise should be able to answer minus the rancor. But instead his response is to attack the driver, InDOT and anyone who questions the pat, predictable answers he gives.
charlie hebdoSo does the railroad have any responsibility for the impact the changes it made in its ROW have on the crossing?
While the short answer is "not really" (and for the reasons MC gave, including the 2' rule) the slightly longer answer is more involved, and perhaps more will in fact be made of it in future.
There are quite a number of these situations in modern practice, where a road initially paralleled a railroad line and has subsequently been widened and opened to heavier traffic or now sees traffic incapable of negotiating the kind of vertical transition possible in the now-existing space. Meanwhile older practice in ballast maintenance has often resulted in elevating the height of the effective prism in these areas over time, probably above what initial drainage required.
It occurs to me that modern ballast-maintenance equipment, and modern practice in aligning and operating it, easily allows for a one-time undercutting to proper grade, perhaps at the same time the drainage is cleared or improved to allow a lowered subgrade. At the same time in most locations PSR (or legitimate operating restrictions on passenger-train speed) might allow more vertical approach curvature to these locations, meaning that only the detailed undercutting near or involving these crossings need be undertaken, and no additional liability or real safety concern would be added by reducing the prism height 'through town'.
In my opinion this is not something the railroads should have to 'subsidize' or that should be required as an unfunded mandate; however, they would derive some theoretical benefit from clean ballast and proper drainage. The key is to make modern complete track replacement more cost-effective here. I note that more than one equipment manufacturer even has YouTube videos showing how their equipment can be used even on branch lines to restore proper grade in what is essentially one pass; if we are to have a 'stimulus' benefiting railroads in a Biden or Harris administration, this seems like a remarkably useful opportunity for 'pump priming' to provide the equipment and training, and then subsidize some large percentage of 'local' or 'state' responsibility as is the case for many improvement projects at present. This would lower the absolute crossing height; local and state would then be fully responsible, as they are at present and in my opinion always should be, for the approach grading to that 2' margin and for appropriate signage (and assumption of full responsibility for all traffic across the crossings).
Lets be a devil advocate. Near here are two crossings that flooded in the past. No way the RR is going to allow lowering the track for the crossing that would prevent trains to operate with water just over tracks.
blue streak 1Lets be a devil's advocate. Near here are two crossings that flooded in the past. No way the RR is going to allow lowering the track for the crossing that would prevent trains to operate with water just over tracks.
I feel like Hammy the squirrel...
Ran up and down the line from the scene of the collision. The next crossing north is even higher - makes the subject crossing look flat.
The next crossing south is pretty much flat.
Continuing further north - the railroad passes over the next road it encounters.
I doubt flooding is the issue - more like trying to level out a short grade.
Larry Resident Microferroequinologist (at least at my house) Everyone goes home; Safety begins with you My Opinion. Standard Disclaimers Apply. No Expiration Date Come ride the rails with me! There's one thing about humility - the moment you think you've got it, you've lost it...
tree68 I feel like Hammy the squirrel... Ran up and down the line from the scene of the collision. The next crossing north is even higher - makes the subject crossing look flat. The next crossing south is pretty much flat. Continuing further north - the railroad passes over the next road it encounters. I doubt flooding is the issue - more like trying to level out a short grade.
The driver was trying to detour around a construction zone. I would opine an asphalt milling operation based upon another video showing a Wirtgen milling machine attempting to pull the semi backwards off the crossing.
For long stretches of this road between Pendleton and Lawrence (Indy), the INDOT R/W and the RR R/W are one and the same. This doesn't leave a lot of room to lessen the vertical grade.
The biggest question I'd like answered - was the emergency notification number called? If not - then we may need to rethink the ENS system and design of the signs.
It's been fun. But it isn't much fun anymore. Signing off for now.
The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.t fun any
Would it not be possible to design a fail safe that would detect a vehicle fouling the tracks and automatically notify any oncoming trains? Maybe even apply the train's brakes without the need for the engineer to do so depending on how close the train is. A phone number on a pole might not be quick enough when every precious second matters.
UlrichWould it not be possible to design a fail safe that would detect a vehicle fouling the tracks and automatically notify any oncoming trains? Maybe even apply the train's brakes without the need for the engineer to do so depending on how close the train is. A phone number on a pole might not be quick enough when every precious second matters.
Equipping locomotives with ESP devices - yep that is the ticket.
Stupid is as stupid does and the stupidity deserves whatever the outcome is.
Never too old to have a happy childhood!
I didn't think it was such a bad idea. It's pretty easy to get hung up on a crossing, more so in bad weather. it's happened to me before.. snow and ice had changed the profile of the road.. pulling a drop frame trailer and BOOM.. just like that I had no millimetres to spare and hung up on a rail I became! Fortunately for me, a snow shovel and a bit of jostling got me released and on my way. But I suppose things could have ended differently. Would a phone number on a pole have helped? Not back then.. smart phones.. even flip phones and those massive brick phones were the realm of science fiction. But I would imagine using a smartphone with gloves on in freezing -40 degree temperatures would have been challenging even then when my fingers were young and nimble. Sensors and an automated stop mechanism makes sense to me..but who knows.. I'm no engineer.
Ulrich I didn't think it was such a bad idea. It's pretty easy to get hung up on a crossing, more so in bad weather. it's happened to me before.. snow and ice had changed the profile of the road.. pulling a drop frame trailer and BOOM.. just like that I had no millimetres to spare and hung up on a rail I became! Fortunately for me, a snow shovel and a bit of jostling got me released and on my way. But I suppose things could have ended differently. Would a phone number on a pole have helped? Not back then.. smart phones.. even flip phones and those massive brick phones were the realm of science fiction. But I would imagine using a smartphone with gloves on in freezing -40 degree temperatures would have been challenging even then when my fingers were young and nimble. Sensors and an automated stop mechanism makes sense to me..but who knows.. I'm no engineer.
Stopping a train in Emergency has a special set of risks. The balance tips in favor of taking those risks if the object to be hit is another train. But taking those risks when the object is just a car, hopefully an empty one, is not good odds.
The crew, and passengers, if it is a passenger train, are at less risk hitting a car than slaming on the trains brakes automaticly when the hazzard may not even exist by the time you get there.
I could be wrong, but I think most of the professional railroaders here will agree.
The answer is smarter humans and better grade crossing construction standards, one being easier to impliment than the other......
Sheldon
ATLANTIC CENTRAL Ulrich I didn't think it was such a bad idea. It's pretty easy to get hung up on a crossing, more so in bad weather. it's happened to me before.. snow and ice had changed the profile of the road.. pulling a drop frame trailer and BOOM.. just like that I had no millimetres to spare and hung up on a rail I became! Fortunately for me, a snow shovel and a bit of jostling got me released and on my way. But I suppose things could have ended differently. Would a phone number on a pole have helped? Not back then.. smart phones.. even flip phones and those massive brick phones were the realm of science fiction. But I would imagine using a smartphone with gloves on in freezing -40 degree temperatures would have been challenging even then when my fingers were young and nimble. Sensors and an automated stop mechanism makes sense to me..but who knows.. I'm no engineer. Stopping a train in Emergency has a special set of risks. The balance tips in favor of taking those risks if the object to be hit is another train. But taking those risks when the object is just a car, hopefully an empty one, is not good odds. The crew, and passengers, if it is a passenger train, are at less risk hitting a car than slaming on the trains brakes automaticly when the hazzard may not even exist by the time you get there. I could be wrong, but I think most of the professional railroaders here will agree. The answer is smarter humans and better grade crossing construction standards, one being easier to impliment than the other...... Sheldon
Ok, so lets make the sensor and the whole shebang a little smarter and a little bit more interactive. Now it senses a tanker truck.. it "interacts" with the approaching train's onboard computer.. together they determine that a brake application of X lbs is optimal to stop the train without it derailing. Oh but hold on!.. the truck has cleared the crossing! The sensors relay that latest bit of information to the train, and just like that the brakes are released and back to normal. Star Wars tech? Not really.. Star Wars was 40 + years ago. I think we've got the technology to do it.. but maybe not! Or maybe it would be way too expensive.. I don't know. But to me it would seem possible to do that.
Ulrich ATLANTIC CENTRAL Ulrich I didn't think it was such a bad idea. It's pretty easy to get hung up on a crossing, more so in bad weather. it's happened to me before.. snow and ice had changed the profile of the road.. pulling a drop frame trailer and BOOM.. just like that I had no millimetres to spare and hung up on a rail I became! Fortunately for me, a snow shovel and a bit of jostling got me released and on my way. But I suppose things could have ended differently. Would a phone number on a pole have helped? Not back then.. smart phones.. even flip phones and those massive brick phones were the realm of science fiction. But I would imagine using a smartphone with gloves on in freezing -40 degree temperatures would have been challenging even then when my fingers were young and nimble. Sensors and an automated stop mechanism makes sense to me..but who knows.. I'm no engineer. Stopping a train in Emergency has a special set of risks. The balance tips in favor of taking those risks if the object to be hit is another train. But taking those risks when the object is just a car, hopefully an empty one, is not good odds. The crew, and passengers, if it is a passenger train, are at less risk hitting a car than slaming on the trains brakes automaticly when the hazzard may not even exist by the time you get there. I could be wrong, but I think most of the professional railroaders here will agree. The answer is smarter humans and better grade crossing construction standards, one being easier to impliment than the other...... Sheldon Ok, so lets make the sensor and the whole shebang a little smarter and a little bit more interactive. Now it senses a tanker truck.. it "interacts" with the approaching train's onboard computer.. together they determine that a brake application of X lbs is optimal to stop the train without it derailing. Oh but hold on!.. the truck has cleared the crossing! The sensors relay that latest bit of information to the train, and just like that the brakes are released and back to normal. Star Wars tech? Not really.. Star Wars was 40 + years ago. I think we've got the technology to do it.. but maybe not! Or maybe it would be way too expensive.. I don't know. But to me it would seem possible to do that.
Trains on Main tracks with authority to move at maximum authorized speed are not line of sight vehicles.
Your tanker gets 'stuck' and initiates 'your technology' when the train is two or three miles from the tanker and the train stops - out of sight of the tanker, that has extracated itself from its situation and had moved on. Now someone has to walk to the location the tanker WAS to see if it is OK for the train to proceed. With PSR size trains is is taking the 2 1/2 to 3 mile long trains well over a mile to bring the train to a SAFE stop from maximum allowed speeds.
Ulrich Would it not be possible to design a fail safe that would detect a vehicle fouling the tracks and automatically notify any oncoming trains?
Would it not be possible to design a fail safe that would detect a vehicle fouling the tracks and automatically notify any oncoming trains?
Crossing signals are required to activate at least 20 seconds before the train arrives. Typically a crossing with modern equipment and a "normal" traffic mix will be configured so that the signals activate about 30 seconds before arrival. A truck that starts across the crossing ~5 seconds after the lights start flashing (allowing for normal reaction times and stopping distance) may only clear the tracks 10-15 seconds before the train arrives. That is way less than the time it takes to stop a train. If you add a system that slows the train every time there is a vehicle on the tracks in front of it, to make sure the train can stop for that vehicle, you would essentially be limiting train speeds to maybe 20 MPH or so.
Now if you GREATLY increase the crossing warning time, then this kind of thing becomes a theoretical possibility. I believe on the Illinois higher-speed rail route they are using something like 80 seconds of warning time, 3-4 times what is normal. But they also have 4-quad gates at all the crossings. If you use 80 seconds of warning time at a 2-quad gate setup, people would get impatient and drive around the gates, and you'd be back to where you started.
Now, maybe you meant to ask whether it is technologically possible to detect if a vehicle is STUCK on the tracks unable to move, and warn trains acordingly? I think the answer is "No," although AI is progressing so fast that's conceivable it could happen some day. Then it will become a cost-benefit question. Train-hits-stuck-vehicle crashes can definitely be dangerous, including to the train crew, but my guess is that the are statistically less likely to result in injuries than any other type of train-vehicle crash. And if you consider lifecycle costs of maintaining such a system, it's might just be cheaper in most cases to fix the road profile.Or use all that AI to build a better truck driver.
Dan
dpeltierNow, maybe you meant to ask whether it is technologically possible to detect if a vehicle is STUCK on the tracks unable to move, and warn trains acordingly? I think the answer is "No"...
In my opinion little more than 'forwarding' video feed and some playback or interface controls to the railroad(s) or companies involved would be necessary; it would be their subsequent responsibility to arrange to determine affected trains and order them to slow or stop or update their electronic 'documentation' to ensure they make a safe and vigilant 'approach'.
Overmod dpeltier Now, maybe you meant to ask whether it is technologically possible to detect if a vehicle is STUCK on the tracks unable to move, and warn trains acordingly? I think the answer is "No"... The answer, technically, is "yes" and the pieces of the solution have been known for some time. The system of cameras used for crossing enforcement can also be used, with comparatively simple machine-vision, to detect vehicles stopped for more than a given time; only slight added capacity would detect attempts to 'free' a stuck vehicle by rocking, or gesticulations made by a driver, or in theory even coded hand signs or recognition of language including agreed key words or phrases
dpeltier Now, maybe you meant to ask whether it is technologically possible to detect if a vehicle is STUCK on the tracks unable to move, and warn trains acordingly? I think the answer is "No"...
The answer, technically, is "yes" and the pieces of the solution have been known for some time. The system of cameras used for crossing enforcement can also be used, with comparatively simple machine-vision, to detect vehicles stopped for more than a given time; only slight added capacity would detect attempts to 'free' a stuck vehicle by rocking, or gesticulations made by a driver, or in theory even coded hand signs or recognition of language including agreed key words or phrases
I said "stuck" on the tracks, not "stopped" on the tracks. Vehicles stop on the tracks all the time without getting stuck. If you assume that a vehicle stopped on the tracks for more than n seconds is stuck, then n is going to have to be very high to get to an acceptable level of false positives. High enough you're still going to risk some collisions. Certainly high enough that someone could just call the dang ENS number and stop trains that way!
As to the other clues you suggest using, I would just point out that these behaviors are all the exact opposite of what someone should be doing in these circumstances, and that many motorists are smart enough not to try them.
Also - if you had to implement such a system today, it would not be based on machine vision. That has been tested for the related application of detecting vehicles occupying a crossing in a 4-quad gate system and found to be too sensitive to lighting and weather conditions. Inductor loops have been the standard for a while now, with a newer radar-based system also in use in some places. (These systems are used for Exit Gate Management; the "exit gates" on the far side of the crossing in the direction of travel don't descend if there's a vehicle occupying the crossing. Note that these systems don't attempt to differentiate between "stopped" and "stuck" vehicles.)
If I recall correctly, either FHWA or OLI had what they considered to be a significant campaign promoting awareness of ENS signs just last year. It doesn't appear to have gotten through to everyone, unfortunately. Maybe something different is needed terms of signage or education, but that's going to produce a far better result than any system we have in the present day for autonomously differentiating between a stopped vehicle and a stuck vehicle.
dpeltierNow, maybe you meant to ask whether it is technologically possible to detect if a vehicle is STUCK on the tracks unable to move, and warn trains acordingly?
CSX recently installed what is reported to be an occupancy detector for the diamond at Deshler. An unassuming device, it sits largely unnoticed in the SW quadrant of the diamond, a few feet from both rails.
Because of the amount of interconnected metal in a diamond, a conventional track circuit is useless there. I think I saw it offered that this was mostly for hi-railers. Most trains will occupy the blocks on both sides.
Nonetheless - If one couples an occupancy detector such as that with the crossing circuit you then get a quarter mile (more or less) of warning that there's something on the crossing when the gates go down.
In this case, however, it appears that the truck fouling the crossing was probably visible way before that, and most trains will require well more than that quarter mile to stop. Such a system would have to also use the track circuits to provide advance warning.
It would be tough to tie that all together so it provided advance warning, but didn't create false warnings.
We also have the problem of crossings without automatic protection at all - more than a few pieces of farm equipment have "bought the farm" at such crossings.
In the end, it goes back to the vehicle operator, the same as the incident in Ashland.
1976.
People who were born in 1976 aren't old, but if you're a man you're likely balding and what hair is left is turning gray. And if you're a women your child bearing years are now behind you. In July of 1976 Time magazine ran a beautiful cover photo of Mars taken from Viking 1 (and then viking 2).. a spacecraft that was landed remotely on the surface of the planet. In 1976! Inside the same issue were numerous color photos.. all of them breathtaking to say the least. What am I leading to? My point is we could do all of this over 44 years ago! And that was long AFTER we'd landed a probe on balmy Venus! Surely bringing technology to bear on the problem of having trucks hung up on crossings without Billy Bumstead having to manually inspect each crossing is within our grasp in 2020..
UlrichSurely bringing technology to bear on the problem of having trucks hung up on crossings without Billy Bumstead having to manually inspect the crossing is within our grasp in 2020..
Such crossing fouling technoloy is certainly within our grasp, and I suspect it will be implemented within ten years. There has been a lot of information published about it being planned. It just monitors each crossing to detect vehicles that are not clear and have been fouling for a period of time longer than the time required for a typical traffic congestion event.
Crossing vehicle stalling is a less likely danger than drivers trying to beat the train, but stalling still occurs often and is worth addressing. Stalling is easier to detect than a driver's intent to take a risk by running the crossing signals.
In most cases of vehicles stalled on the track, there is ample time to stop any train if they are warned when the stalling first occurs. So there is no need for braking with an emergency application.
Stalling is probably more likely with trucks than with passenger vehicles. There is more risk to the train crew in hitting a truck than with passenger vehicles. So the railroad company has an interest in preventing collisions with stalled vehicles.
Perhaps the Pinto Principle applies here...
dpeltierI said "stuck" on the tracks, not "stopped" on the tracks.
Certainly my experience with machine vision, including the improvements being made for autonomous vehicles, makes your supposed issues with 'lighting and weather' obsolescent at best. The point of the camera-based system is precisely that it has alternative uses for crossing enforcement and monitoring which 'radar' or 'inductive loops' don't even remotely approximate, the former also involving RF issues, the latter involving cost to position and calibrate and then to keep maintained and powered.
I trust that you have developed an appropriate bridge from crossing circuits to 'third-party' notification systems. Please describe the approach and protocols in detail.
The aftermath, from the Pendelton/Fall Creek Fire Department Facebook Page:
Overmod dpeltier I said "stuck" on the tracks, not "stopped" on the tracks. That's a distinction without a difference, the way I used the term. Almost by definition if a vehicle is 'stopped' for more than a reasonable time, or for no apparent reason related to traffic, it would be considered 'stuck' as far as sending advance warning to approaching trains. Certainly my experience with machine vision, including the improvements being made for autonomous vehicles, makes your supposed issues with 'lighting and weather' obsolescent at best. The point of the camera-based system is precisely that it has alternative uses for crossing enforcement and monitoring which 'radar' or 'inductive loops' don't even remotely approximate, the former also involving RF issues, the latter involving cost to position and calibrate and then to keep maintained and powered. I trust that you have developed an appropriate bridge from crossing circuits to 'third-party' notification systems. Please describe the approach and protocols in detail.
dpeltier I said "stuck" on the tracks, not "stopped" on the tracks.
That's a distinction without a difference, the way I used the term. Almost by definition if a vehicle is 'stopped' for more than a reasonable time, or for no apparent reason related to traffic, it would be considered 'stuck' as far as sending advance warning to approaching trains.
It's good to see some folks realizing there is a problem and there must be better ways to deal with it.
charlie hebdo Overmod dpeltier I said "stuck" on the tracks, not "stopped" on the tracks. That's a distinction without a difference, the way I used the term. Almost by definition if a vehicle is 'stopped' for more than a reasonable time, or for no apparent reason related to traffic, it would be considered 'stuck' as far as sending advance warning to approaching trains. Certainly my experience with machine vision, including the improvements being made for autonomous vehicles, makes your supposed issues with 'lighting and weather' obsolescent at best. The point of the camera-based system is precisely that it has alternative uses for crossing enforcement and monitoring which 'radar' or 'inductive loops' don't even remotely approximate, the former also involving RF issues, the latter involving cost to position and calibrate and then to keep maintained and powered. I trust that you have developed an appropriate bridge from crossing circuits to 'third-party' notification systems. Please describe the approach and protocols in detail. It's good to see some folks realizing there is a problem and there must be better ways to deal with it.
I proposed a pretty straight forward solution earlier, maybe in the other thread about the car, smarter humans and better grade crossing construction standards, again, one being considerably easier to effect than the other.
You get more of what you subsidize and less of what you tax. Maybe we should reconsider what we subsidize? And what we tax?
Overmod dpeltier I said "stuck" on the tracks, not "stopped" on the tracks. That's a distinction without a difference, the way I used the term. Almost by definition if a vehicle is 'stopped' for more than a reasonable time, or for no apparent reason related to traffic, it would be considered 'stuck' as far as sending advance warning to approaching trains.
Only if you don't care about the effect of false positives on train traffic, or if you assume that most vehicles stopped for no reason on the tracks will in fact still be there when a train arrives. I don't know if there is data to support this assumption. Anecdotally, I have seen vehicles stopped on railroad tracks that eventually moved off the tracks with no assistance (and without the benefit of an interconnected traffic light), so I don't know why you would think they don't exist.
Overmod Certainly my experience with machine vision, including the improvements being made for autonomous vehicles, makes your supposed issues with 'lighting and weather' obsolescent at best.
Certainly my experience with machine vision, including the improvements being made for autonomous vehicles, makes your supposed issues with 'lighting and weather' obsolescent at best.
They're not "my" supposed issues, they are issues that have been identified by the literature (see e.g. Medina, Chitturi, and Benekohal, "Effects of fog, snow, and rain on video detection systems at intersections", Transportation Letters, 2:1, 1-12, 2013). And regardless of whether they are being addressed by newer technology, they largely explain why the only technologies currently used for vehicle detection are rail crossings are loops and radar. As I said at the outset, some day there will probably be technology that would allow one to detect stuck vehicles at a crossing, and then deployment will depend mostlu on cost and benefit. But we're not there today.
tree68 The aftermath, from the Pendelton/Fall Creek Fire Department Facebook Page:
Thanks to Chris / CopCarSS for my avatar.
Murphy SidingLarry- Something doesn't look right about that picture. Why is the fireman standing that close to a fireball?
Aside from the tires, there's little explosion hazard there. It's little more than a big bonfire of burning plastic and foam.
Besides, something we try to do is a "360" assessment of the incident. He's taking a look at what needs to be done, and how. He's also in full turnouts and on air.
Our community is FREE to join. To participate you must either login or register for an account.