zugmann Was the engineer running the train - or was the software?
Was the engineer running the train - or was the software?
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
BigJim Euclid So when this train applied dynamic braking of the two head end locomotives, it only retarded the wheels of those two locomotives. All of the 210 cars trailing the locomotives had no braking at all acting on their wheels. If you use air brakes, then all 840 car wheels of the train will have brake shoes pressing into them. Once again, you have no idea how to run a train nor the proper use of the automatic brake!
Euclid So when this train applied dynamic braking of the two head end locomotives, it only retarded the wheels of those two locomotives. All of the 210 cars trailing the locomotives had no braking at all acting on their wheels. If you use air brakes, then all 840 car wheels of the train will have brake shoes pressing into them.
Once again, you have no idea how to run a train nor the proper use of the automatic brake!
If the FRA is right about part of the cause being the train handling, what might they be referring to?
EuclidSo when this train applied dynamic braking of the two head end locomotives, it only retarded the wheels of those two locomotives. All of the 210 cars trailing the locomotives had no braking at all acting on their wheels. If you use air brakes, then all 840 car wheels of the train will have brake shoes pressing into them.
.
tree68 Overmod charlie hebdo You missed the key point of Bucky's post, namely that the doomed train lacked sufficient braking capability. But wasn't a key point at Springfield that there was too much (or more precisely too quickly, or poorly modulated) head-end-only dynamic braking, causing run-in acceleration in part of the lightly-loaded intermediate section of the consist? One could argue that there was too much braking capacity, in the wrong place...
Overmod charlie hebdo You missed the key point of Bucky's post, namely that the doomed train lacked sufficient braking capability. But wasn't a key point at Springfield that there was too much (or more precisely too quickly, or poorly modulated) head-end-only dynamic braking, causing run-in acceleration in part of the lightly-loaded intermediate section of the consist?
charlie hebdo You missed the key point of Bucky's post, namely that the doomed train lacked sufficient braking capability.
But wasn't a key point at Springfield that there was too much (or more precisely too quickly, or poorly modulated) head-end-only dynamic braking, causing run-in acceleration in part of the lightly-loaded intermediate section of the consist?
One could argue that there was too much braking capacity, in the wrong place...
An thus we have the most critical element of a Engineers performance in getting his train SAFELY across the road.
Gross manipulation of throttle and braking systems can tear trains apart and derail them.
Engineers are ones who MASTER the finess of manipulating the tools of their trades. How do Engineers master their tools, train after train after train, trip after trip after trip - every train and trip is a learning experience, several synapse of gray matter are filed about the minuate of today's train operation as compared to hundreds if not thousands of trips with other trains and the sensations those train transmitted to the Engineer when various train handling techniques were applied at various locations on each trip.
Engineers are planning their control inputs miles in advance of when those inputs are going to be necessary, as in many cases certain actions have to be completed prior to those inputs. As operators of automobiles, we are able to safely respond to virtually any situation we can see before the situation becomes an accident. Railroad Engineers on Main tracks cannot operate at track speeds within their range of vision; they have to be planning actions for situations they know will present themselves but they cannot see them yet.
Never too old to have a happy childhood!
Euclid If you use air brakes, then all 840 car wheels of the train will have brake shoes pressing into them.
And all 840 brake shoes are just that much closer to replacement. Brake shoes cost money, as does the labor to replace them. It may not sound like much, but with hundreds of thousands of cars in circulation, it adds up.
Smart automobile drivers use "dynamic braking" all the time. Better known as engine braking. Either that, or let geography help - going up a hill? Just let gravity slow you down. Why wear out one's brakes?
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...
charlie hebdoYou missed the key point of Bucky's post, namely that the doomed train lacked sufficient braking capability.
You missed the key point of Bucky's post, namely that the doomed train lacked sufficient braking capability.
EuclidAt the time of the derailment, the entire 210-car train was being slowed with dynamic braking provided by only by the two units on the head end.
zugmann Euclid It is true that any engineer can derail their train with bad train handling. But I don’t think this fact can rule out the possibility that the longer trains have higher in-train forces that can make it easier for a careless engineer to cause a derailment due in part to poor train handling. It has not even been established that what is considered proper train handling for moderate train lengths may be inadequate for the ultra-long trains. Kind of forgetting trip optimizer is a thing.
Euclid It is true that any engineer can derail their train with bad train handling. But I don’t think this fact can rule out the possibility that the longer trains have higher in-train forces that can make it easier for a careless engineer to cause a derailment due in part to poor train handling. It has not even been established that what is considered proper train handling for moderate train lengths may be inadequate for the ultra-long trains.
Kind of forgetting trip optimizer is a thing.
And in 'Company Eyes' Trip Optimizer is infailible. Even when Trip Optimizer makes train handling actions that cause derailments. Only human engineers make train handling mistakes in the eyes of the company.
EuclidIt is true that any engineer can derail their train with bad train handling. But I don’t think this fact can rule out the possibility that the longer trains have higher in-train forces that can make it easier for a careless engineer to cause a derailment due in part to poor train handling. It has not even been established that what is considered proper train handling for moderate train lengths may be inadequate for the ultra-long trains.
BaltACD Euclid Overmod Why didn't you put 5 and 6 in red, too? The latter is just what you were calling for in the last few weeks. Here is the relevant reference in the eCFR: https://www.ecfr.gov/current/title-49/subtitle-B/chapter-II/part-217/subpart-A/section-217.9 Note subpart (c)(1). I know it can be irritating to wade through the Federal verbiage and cross-reference their priorities with actual railroad reality, but this actually calls for operations testing and not 'more detailed inspections'. I don’t mean to diminish the role of testing. I think all of the points on the FRA list are relevant. It is just that #4 interests me most because it is wrapped around the factors that lead to excess in-train forces. Until that issue is fully understood, it will not be solvable. And to be fully understood, it needs to be tested-sensed on monster trains. Of course sensing the drawbar loading on various mixes of trains that are over 200 cars long will be an awful chore. But if it is done, I suspect it will disclose some surprising slack behavior. Maybe in lieu of measuring forces with sensors to start, they could develop a program that could analyze a hypothetical long train just according to its makeup and location on the line. Then use that program to predict whether the NS train that derailed in Springfield, OH would derail as it did. But I would not stop there, especially if the program says the Springfield derailment would not have happened. Improper, ham fisted, locomotive operation can cause a derailment of virtually any size train - big or small. Cause of the Springfield, OH derailment has not been published - officially or unoffically. Without access to the full investigation of the incident we don't know anything factual. Without factual knowledge we are just making WAGs.
Euclid Overmod Why didn't you put 5 and 6 in red, too? The latter is just what you were calling for in the last few weeks. Here is the relevant reference in the eCFR: https://www.ecfr.gov/current/title-49/subtitle-B/chapter-II/part-217/subpart-A/section-217.9 Note subpart (c)(1). I know it can be irritating to wade through the Federal verbiage and cross-reference their priorities with actual railroad reality, but this actually calls for operations testing and not 'more detailed inspections'. I don’t mean to diminish the role of testing. I think all of the points on the FRA list are relevant. It is just that #4 interests me most because it is wrapped around the factors that lead to excess in-train forces. Until that issue is fully understood, it will not be solvable. And to be fully understood, it needs to be tested-sensed on monster trains. Of course sensing the drawbar loading on various mixes of trains that are over 200 cars long will be an awful chore. But if it is done, I suspect it will disclose some surprising slack behavior. Maybe in lieu of measuring forces with sensors to start, they could develop a program that could analyze a hypothetical long train just according to its makeup and location on the line. Then use that program to predict whether the NS train that derailed in Springfield, OH would derail as it did. But I would not stop there, especially if the program says the Springfield derailment would not have happened.
Overmod Why didn't you put 5 and 6 in red, too? The latter is just what you were calling for in the last few weeks. Here is the relevant reference in the eCFR: https://www.ecfr.gov/current/title-49/subtitle-B/chapter-II/part-217/subpart-A/section-217.9 Note subpart (c)(1). I know it can be irritating to wade through the Federal verbiage and cross-reference their priorities with actual railroad reality, but this actually calls for operations testing and not 'more detailed inspections'.
Why didn't you put 5 and 6 in red, too? The latter is just what you were calling for in the last few weeks.
Here is the relevant reference in the eCFR:
https://www.ecfr.gov/current/title-49/subtitle-B/chapter-II/part-217/subpart-A/section-217.9
Note subpart (c)(1).
I know it can be irritating to wade through the Federal verbiage and cross-reference their priorities with actual railroad reality, but this actually calls for operations testing and not 'more detailed inspections'.
Improper, ham fisted, locomotive operation can cause a derailment of virtually any size train - big or small.
Cause of the Springfield, OH derailment has not been published - officially or unoffically. Without access to the full investigation of the incident we don't know anything factual. Without factual knowledge we are just making WAGs.
BaltACDCSX has a group of 'box cars' that have been outfitted with geometry measurement equipment that they operate on various regularly scheduled trains operating over the network.
Generally easily spotted by the huge billboard lettering "DO NOT HUMP" on both sides...
At night you can sometimes spot the lasers under the cars.
rdamonTetra Tech’s RailAI system offers the most technologically advanced in revenue service, at track speed, fully autonomous track inspection assessment system available. https://railai.tetratech.com/
https://railai.tetratech.com/
CSX has a group of 'box cars' that have been outfitted with geometry measurement equipment that they operate on various regularly scheduled trains operating over the network. The also have specific locomotives that have been outfitted for track geometry measurements. I know the locomotives wirelessly transmit the exceptions they find to the MofW Desk in Jacksonville who in turn notify the Roadmaster of the affected territory for them to inspect and recitfy. I don't know of the box cars are wirelessly reporting the results of the data they develop.
I know the locomotives were reporting data before I retired in 2016. I would imagine improvements have been made in the past seven years.
Tetra Tech’s RailAI system offers the most technologically advanced in revenue service, at track speed, fully autonomous track inspection assessment system available.
Euclid4. Identify changes to crew training, train handling procedures, train makeup, DPU requirements, limitations to length or tonnage, speed restrictions, track, mechanical, and brake inspection and maintenance requirements necessary to ensure safe operations of longer trains.
BaltACD Euclid BaltACD Euclid Yes I understand that trains change their consists by making setouts and pickups along the way. Why is that a problem? The system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route. A key point here is that the physical train does not have to even exist to be inspected by this system program. All it needs to “inspect” the train is a list of its cars. This is basically what the FRA intends to do in order to solve a problem of dangerously high in-train forces that they believe exists with this new generation of ultra-long trains. The solution is essentially the same as the train marshalling programs that have been developed and sometimes used over the last 15 years. But the FRA finds evidence that the risk of this problem of excess in-train forces has grown larger precisely because of the massive increase in train length in recent years. So they are calling for new research to find and predict these higher in-train forces. Why would anybody besides railroad management oppose that? Your protestations remind me of when a change was made to the National T&E contract that permitted cars for a 'block' to be in other locations in trains and that crews could be required to switch out those 'outliers' with no penalty payments being owed to the crew. Some 'Terminal Management' thought buckshotting cars in the trains was the way to minimize the terminals crew time in building trains. All was well and good until senior division management started asking why it was take three, four or more hours for a train to make a 25 cars set off ON LINE OF ROAD. When it comes to block swapping - a block that is made correctly for Train 1 can be a issue for Train 2, especially when HAZMAT is part of any of the trains. You and Don say the in-train force detection system I propose has already been done. And you also seem to be saying it won’t work because of the block problem you are describing. Can you explain this? If the system cannot work, why was it built? P S R
Euclid BaltACD Euclid Yes I understand that trains change their consists by making setouts and pickups along the way. Why is that a problem? The system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route. A key point here is that the physical train does not have to even exist to be inspected by this system program. All it needs to “inspect” the train is a list of its cars. This is basically what the FRA intends to do in order to solve a problem of dangerously high in-train forces that they believe exists with this new generation of ultra-long trains. The solution is essentially the same as the train marshalling programs that have been developed and sometimes used over the last 15 years. But the FRA finds evidence that the risk of this problem of excess in-train forces has grown larger precisely because of the massive increase in train length in recent years. So they are calling for new research to find and predict these higher in-train forces. Why would anybody besides railroad management oppose that? Your protestations remind me of when a change was made to the National T&E contract that permitted cars for a 'block' to be in other locations in trains and that crews could be required to switch out those 'outliers' with no penalty payments being owed to the crew. Some 'Terminal Management' thought buckshotting cars in the trains was the way to minimize the terminals crew time in building trains. All was well and good until senior division management started asking why it was take three, four or more hours for a train to make a 25 cars set off ON LINE OF ROAD. When it comes to block swapping - a block that is made correctly for Train 1 can be a issue for Train 2, especially when HAZMAT is part of any of the trains. You and Don say the in-train force detection system I propose has already been done. And you also seem to be saying it won’t work because of the block problem you are describing. Can you explain this? If the system cannot work, why was it built?
BaltACD Euclid Yes I understand that trains change their consists by making setouts and pickups along the way. Why is that a problem? The system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route. A key point here is that the physical train does not have to even exist to be inspected by this system program. All it needs to “inspect” the train is a list of its cars. This is basically what the FRA intends to do in order to solve a problem of dangerously high in-train forces that they believe exists with this new generation of ultra-long trains. The solution is essentially the same as the train marshalling programs that have been developed and sometimes used over the last 15 years. But the FRA finds evidence that the risk of this problem of excess in-train forces has grown larger precisely because of the massive increase in train length in recent years. So they are calling for new research to find and predict these higher in-train forces. Why would anybody besides railroad management oppose that? Your protestations remind me of when a change was made to the National T&E contract that permitted cars for a 'block' to be in other locations in trains and that crews could be required to switch out those 'outliers' with no penalty payments being owed to the crew. Some 'Terminal Management' thought buckshotting cars in the trains was the way to minimize the terminals crew time in building trains. All was well and good until senior division management started asking why it was take three, four or more hours for a train to make a 25 cars set off ON LINE OF ROAD. When it comes to block swapping - a block that is made correctly for Train 1 can be a issue for Train 2, especially when HAZMAT is part of any of the trains.
Euclid Yes I understand that trains change their consists by making setouts and pickups along the way. Why is that a problem? The system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route. A key point here is that the physical train does not have to even exist to be inspected by this system program. All it needs to “inspect” the train is a list of its cars. This is basically what the FRA intends to do in order to solve a problem of dangerously high in-train forces that they believe exists with this new generation of ultra-long trains. The solution is essentially the same as the train marshalling programs that have been developed and sometimes used over the last 15 years. But the FRA finds evidence that the risk of this problem of excess in-train forces has grown larger precisely because of the massive increase in train length in recent years. So they are calling for new research to find and predict these higher in-train forces. Why would anybody besides railroad management oppose that?
Your protestations remind me of when a change was made to the National T&E contract that permitted cars for a 'block' to be in other locations in trains and that crews could be required to switch out those 'outliers' with no penalty payments being owed to the crew.
Some 'Terminal Management' thought buckshotting cars in the trains was the way to minimize the terminals crew time in building trains. All was well and good until senior division management started asking why it was take three, four or more hours for a train to make a 25 cars set off ON LINE OF ROAD.
When it comes to block swapping - a block that is made correctly for Train 1 can be a issue for Train 2, especially when HAZMAT is part of any of the trains.
zugmann Euclid The system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route. There's a lot of these models already out there. They will continue to be be refined, and use expanded, I'm sure - but to imply they currently do not exist is disingenious at best.
Euclid The system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route.
There's a lot of these models already out there. They will continue to be be refined, and use expanded, I'm sure - but to imply they currently do not exist is disingenious at best.
We now have a thing now called UP Vision. One of the features is that it lists places of risk areas for draft/buff force. It gives the mile post and even gives the car number where it's calculated the forces will be the most extreme for that location.
I checked a manifest scheduled with 196 loads, 34 empties 25700 tons and 13900 feet in length. There are 25 locations of medium or high (mostly medium, only about 4 high locations) of buff/draft force risks on four different subdivision.
The programs are out there. If they are actually used to rearrange a consist at an originating terminal, I wouldn't know. I think most of the yard masters just use the regular computer programs to "train set" a consist. If the computer doesn't spit out a warning and sets the consist, it's good to go.
Jeff
EuclidYes I understand that trains change their consists by making setouts and pickups along the way. Why is that a problem? The system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route. A key point here is that the physical train does not have to even exist to be inspected by this system program. All it needs to “inspect” the train is a list of its cars. This is basically what the FRA intends to do in order to solve a problem of dangerously high in-train forces that they believe exists with this new generation of ultra-long trains. The solution is essentially the same as the train marshalling programs that have been developed and sometimes used over the last 15 years. But the FRA finds evidence that the risk of this problem of excess in-train forces has grown larger precisely because of the massive increase in train length in recent years. So they are calling for new research to find and predict these higher in-train forces. Why would anybody besides railroad management oppose that?
EuclidThe system I refer to will warn of dangerous train consist/make-up, not only prior to departure, but also with every change in consist along the route.
BaltACD oltmannd BaltACD oltmannd Euclid This issue is not just a matter of having computers and doing modeling. Everybody is doing that every day. The point is that the computer modeling needed for solving the problem of excess in-train forces has got to be directed at that problem. So the first step is to find examples of that problem by measuring in-train forces. Then use the most applicable computer analysis to see if it verifies that same problem. If it does not verify the problem, then try to figure out why a real problem is not detected by the computer program. In other words, you need to test real trains to learn how to build the program that can flag that problem before it causes a derailment. Validation work had long been done. Models work very well. You can't practically model every train for every operational scenario before you dispatch. Train forces are the results of train handling which depend largely on what's happening where on the railroad plus unintended stuff. You gonna model the train going into emergency every 50 feet? For a separated air hose at every one of 200 locations on the train? Or even every combination of DB and service brake application for every spot the train might have to slow down? Not happening. Can't happen. Euc wants all that modeling done for each train that departs its originating terminal - for each and every train, every trip. Probably ending in discipline for somebody if the train derails for any reason anywhere. Yup. And what are you going to use for train handling inputs? You put that exact train out on the road for 10,000 days in a row and you'll have 10,000 unique sets of train handling data (brake, throttle, DB, etc.) Which is a concept Euc can't or won't comprehend. My favorite interest is automobile racing - virtually all forms from F1 through IndyCar, NASCAR and on down to Club racing and the local bull ring dirt tracks. In many cases there are 'spec classes' where the intent of the specifications are to make the cars as identical as humanly possible to that the driver is the variable to the success or failure in any competition. The drivers is the 'master sensor' and is the one that inputs throttle, brake and steering inputs to make the car go - good drivers are faster than less than good driver. I won't hazard any guess to what a 'computer' driver's performance would be. I have not been on any trains that are being operated with the various locomotive operation computer application, as such I am not in any position to comment on how well those programs peform, what their sensory inputs are and how they react to those sensory inputs. Jeff Hergert would be a much better source on their operation and quirks.
oltmannd BaltACD oltmannd Euclid This issue is not just a matter of having computers and doing modeling. Everybody is doing that every day. The point is that the computer modeling needed for solving the problem of excess in-train forces has got to be directed at that problem. So the first step is to find examples of that problem by measuring in-train forces. Then use the most applicable computer analysis to see if it verifies that same problem. If it does not verify the problem, then try to figure out why a real problem is not detected by the computer program. In other words, you need to test real trains to learn how to build the program that can flag that problem before it causes a derailment. Validation work had long been done. Models work very well. You can't practically model every train for every operational scenario before you dispatch. Train forces are the results of train handling which depend largely on what's happening where on the railroad plus unintended stuff. You gonna model the train going into emergency every 50 feet? For a separated air hose at every one of 200 locations on the train? Or even every combination of DB and service brake application for every spot the train might have to slow down? Not happening. Can't happen. Euc wants all that modeling done for each train that departs its originating terminal - for each and every train, every trip. Probably ending in discipline for somebody if the train derails for any reason anywhere. Yup. And what are you going to use for train handling inputs? You put that exact train out on the road for 10,000 days in a row and you'll have 10,000 unique sets of train handling data (brake, throttle, DB, etc.)
BaltACD oltmannd Euclid This issue is not just a matter of having computers and doing modeling. Everybody is doing that every day. The point is that the computer modeling needed for solving the problem of excess in-train forces has got to be directed at that problem. So the first step is to find examples of that problem by measuring in-train forces. Then use the most applicable computer analysis to see if it verifies that same problem. If it does not verify the problem, then try to figure out why a real problem is not detected by the computer program. In other words, you need to test real trains to learn how to build the program that can flag that problem before it causes a derailment. Validation work had long been done. Models work very well. You can't practically model every train for every operational scenario before you dispatch. Train forces are the results of train handling which depend largely on what's happening where on the railroad plus unintended stuff. You gonna model the train going into emergency every 50 feet? For a separated air hose at every one of 200 locations on the train? Or even every combination of DB and service brake application for every spot the train might have to slow down? Not happening. Can't happen. Euc wants all that modeling done for each train that departs its originating terminal - for each and every train, every trip. Probably ending in discipline for somebody if the train derails for any reason anywhere.
oltmannd Euclid This issue is not just a matter of having computers and doing modeling. Everybody is doing that every day. The point is that the computer modeling needed for solving the problem of excess in-train forces has got to be directed at that problem. So the first step is to find examples of that problem by measuring in-train forces. Then use the most applicable computer analysis to see if it verifies that same problem. If it does not verify the problem, then try to figure out why a real problem is not detected by the computer program. In other words, you need to test real trains to learn how to build the program that can flag that problem before it causes a derailment. Validation work had long been done. Models work very well. You can't practically model every train for every operational scenario before you dispatch. Train forces are the results of train handling which depend largely on what's happening where on the railroad plus unintended stuff. You gonna model the train going into emergency every 50 feet? For a separated air hose at every one of 200 locations on the train? Or even every combination of DB and service brake application for every spot the train might have to slow down? Not happening. Can't happen.
Euclid This issue is not just a matter of having computers and doing modeling. Everybody is doing that every day. The point is that the computer modeling needed for solving the problem of excess in-train forces has got to be directed at that problem. So the first step is to find examples of that problem by measuring in-train forces. Then use the most applicable computer analysis to see if it verifies that same problem. If it does not verify the problem, then try to figure out why a real problem is not detected by the computer program. In other words, you need to test real trains to learn how to build the program that can flag that problem before it causes a derailment.
Validation work had long been done. Models work very well.
You can't practically model every train for every operational scenario before you dispatch. Train forces are the results of train handling which depend largely on what's happening where on the railroad plus unintended stuff.
You gonna model the train going into emergency every 50 feet? For a separated air hose at every one of 200 locations on the train? Or even every combination of DB and service brake application for every spot the train might have to slow down?
Not happening. Can't happen.
Euc wants all that modeling done for each train that departs its originating terminal - for each and every train, every trip. Probably ending in discipline for somebody if the train derails for any reason anywhere.
Yup. And what are you going to use for train handling inputs? You put that exact train out on the road for 10,000 days in a row and you'll have 10,000 unique sets of train handling data (brake, throttle, DB, etc.)
Which is a concept Euc can't or won't comprehend.
My favorite interest is automobile racing - virtually all forms from F1 through IndyCar, NASCAR and on down to Club racing and the local bull ring dirt tracks. In many cases there are 'spec classes' where the intent of the specifications are to make the cars as identical as humanly possible to that the driver is the variable to the success or failure in any competition. The drivers is the 'master sensor' and is the one that inputs throttle, brake and steering inputs to make the car go - good drivers are faster than less than good driver. I won't hazard any guess to what a 'computer' driver's performance would be.
I have not been on any trains that are being operated with the various locomotive operation computer application, as such I am not in any position to comment on how well those programs peform, what their sensory inputs are and how they react to those sensory inputs. Jeff Hergert would be a much better source on their operation and quirks.
Our community is FREE to join. To participate you must either login or register for an account.