After this weekend's NY train crash I started wondering if commuter trains that are being pushed by an engine are more likely to derail than trains that are being pulled by an engine. My wife takes a commuter train to work each day and they run as a "puller" in the morning and a "pusher" in the evening so I am curious although I doubt an answer would lead to anything making her journey safer.
Are the trains run that way simply because it's cheaper - no turnaround is needed?
My very limited experience with pushing things rather than pulling them leads me to conclude that pushing is more dangerous, but I was wondering if anyone out there knew for sure or at least knew where to ask.
As a physics problem, it's more of an issue pushing than pulling, yes.
Operated within the defined boundaries and with maintained roadbeds, I doubt there's a statistical difference observable. OTOH, many (most??) accidents have a factor of human error as, for example, it has been suggested the above _may_ have been operating above the set speed limits altho I've certainly not seen any corroborating evidence as yet.
The place where there are data if it exits is at the NTSB altho I suspect the level of detail for the question would require more research than simply looking up tabulated statistics that are available online or for download.
They've been both pushing and pulling cars since the beginning of RR days some
170 years ago. So you may assume that the safety considerations are well understood by now (whether followed or not is another matter). My guess is that pulling is always better if only for the fact that the engineer can see better. But that does not matter - sometimes there's just no way to switch the engine, so they have to be prepared to move it both ways anyhow.
The Metro-North trains are motor coach cars. The traction motors are on the passenger "coach" car itself. There is no separate locomotive. They neither push nor pull.
My assumptions align with yours (pushing a rope is hard) but the only "proof" I have is that trains typically pull. I sat waiting for a 130 car freight train, pulled (slowly) by four locomotives, on the way to work this morning.
HO trains don't like being pushed very well around curves. I think it will be blamed on the speed. Do they have CC TV from the front car to the locomotive in the rear or are they blindly pushing?
On 12/2/2013 11:40 AM, snipped-for-privacy@attt.bizz wrote: ...
Although depending on grade and loading, they often add "helpers" in the middle and/or at the rear as well.
"Helper", UT is aptly named--it's the location the D&RGW stationed additional engines to add to trains climbing Soldier Summit heading to Salt Lake. It's in Carbon County NW of Price. The county name gives a real hint as to the loading of most of the trains heading out...
Awhile back, I heard of a wreck in Japan. Seems the fairly new train operator miscalculated a stop at a station. And then ran late. He tried to make up for it by zooming, but lost his train on a curve. Aparently, the Japanese concept of training is scrubbing tracks with tooth brush, and lots of yelling at.
Why? Look at the forces, centers of mass, and you can see that a sideways thrust to a pulled train has smaller effect than a sideway thrust to a pushed train. The pushing forces exacerbate the offsets.
From personal experience at moving heavy loads by pulling [preferred rarely waste energy in sideways motion] and by pushing [hate it, the item often veers off to the side]
However, if exactly straight/inline, push or pull should be no difference. I'm just envisioning what happens when something isn't quite right. you're pushing straight and the item wants to slightly veer off line, then your pushing just made it veer off more. like balancing a ball on a knife edge, very difficult to remain stable.
Now you've gone and done it. You've dared to disagree and claim that something krw posted is wrong. I'm sure you'll be having plenty of insults hurled your way shortly. Then you'll be accused of being a lefty. For the record, I believe you are correct and this latest gaff will be added to krw's long running list.
It looks like there is one in the pictures, too, though it might be a motor coach pushing "dummies", too. The pictures aren't very clear. The section of track from Croton-Harmon down to Grand Central Station is electrified - diesels aren't used, so "locomotive" doesn't sound right.
There's still something wrong with the description, sorta like the "automatic shotgun".
Someone suggested his HO train set as an example so I went next door where my neighbor has his trains set up to run some experiments. Pushing seems guaranteed to have a worse outcome for what seems to be a simple reason. If the lead car hits something or encounters bent tracks, a pushed train is going to accordian those cars from behind. A pulled train doesn't have the weight of the locomotive *behind* the string of cars adding force to a collapse.
The cars in the NY state incident, from what I could see, were really flung far off the track. While I realize that in a pulled situation they could have "snapped the whip" and rolled just as far away from the rails as in a pushed train but this incident seemed to have the cars well-scattered.
It will be interesting to see how the investigation unfolds. It seems to be reminiscent of the issues involved between front wheel and rear wheel drive automobiles.
What I am trying to get at is even considering human error, are the outcomes of the mistakes statistically different for pushed v. pulled trains?
There are all sorts of way the NTSB can calculate train speed now, from satelites to transponders to even examining the GPS enabled phones and cell tower records of the passengers. (Or in one sad case here in DC, the train operator's cell phone information as she was on the phone yakking when she killed herself and a number of passengers by plowing into the train in front - a real "push" accident).
I haven't done any follow-up reading or viewing since this morning but I assume it's going to take at least a few days for the preliminary work to be done. No doubt there will be a tremendous amount of speculation occurring in the meantime. (-" Some even right here!
There might already be some sort of analysis of the problem. Insurers obsess over things like this, if only to charge higher premiums if their actuaries can demonstrate a statistically valid increase in risk for certain processes/procedures.
Thanks for your input, dpb. Plenty of things to consider . . .
No, that's not the case. You're thinking of electrically-powered multiple-unit cars.
The train involved had unpowered coaches with a locomotive on the rear end and a "cab control car" at the opposite end (which in this instance was the front of the train). The cab car has a throttle and brake setup and the control circuits are carried through the coaches in the middle and back to a jumper connection on the engine.
Back when I was running trains, I never liked the cab control setup, but that was just me. First, being an engineman, I wanted to BE "on the engine" to better know what it was doing (you don't get enough feedback on the other end when the engine is pushing). Secondly, the ride is just rougher in push mode. Easier to control how the train handles when the engine is pulling.
Thanks for input, that's what I thought I read in the brief initial reports.
The more I played with train cars tonight the more I believe that pushing is probably going to turn out to be statistically higher in deaths per mile than pulling. Pushed cars tend to sway out in response to force from behind. Pulled cars have a much higher resistance to jack-knifing.
It's pretty remarkably easy to demonstrate with toy train cars. If the lead car of a pushed train derails, the cars behind will jack-knife. A heavy locomotive might easily dislodge an obstruction on the track like a car whereas an aluminum commuter train car acting as the lead might not.
Whatever the cause, it's going to be an interesting investigation. In this day and age of instant computerized control, excessive speed should be caught pretty early and possibly automatically. I remember that after years of driving little sports cars I switched to a minivan and realized, in white-knuckled, feeling the van lift up slightly panic, that I couldn't take curves at anywhere NEAR the speed that I used to in a little Honda. (-:
OTOH, my occasional race car driving wife says "if your rear wheels don't "drift" through a turn, you're not going fast enough." With all these automatic control and braking systems it's hard to believe, given the less-than-stellar record of at least a few train drivers, that some automatic system would NOT put on the brakes before entering a curve at a speed guaranteed to cause a derailment.
I've been playing around with 027 trains but the same thing is true of them. If they derail or hit an obstruction, the cars behind will accordian outwards, or worse, telescope into each other.
Operator error. Ask any commercial pilot how blame is apportioned, if only for legal reasons.
I believe they have a control cab upfront that is connected to the pushing engine. From what little reading I've done so far, the configurations are many. The key factor seems to be that turning a train around requires a lot of real estate or a special roundhouse and these push/pull and other configurations eliminate the need to reverse the train.
I wonder if they have footage of the train derailing? More and more CCTV camera are being installed everywhere - they had an amazing amount of meteor footage from Russia that came from parking lot cameras, dashcams and even cell phones. There was a fascinating TV show about how they collected all those videos and were able to determine the meteorite's trajectory. It showed that it came directly out of the solar glare and was invisible to the asteroid detection cameras/satelites now in use because they are "blinded" by the sun.
Doubt there's sufficient data to make a statistically valid conclusion on that factor irrespective of the confounding factors of so many other variables being as there's no control in the "experiment" to allow comparison of the one effect.
...
The NTSB announced from black box data something like 82 mph just ahead of a 30 mph-rated curve. They've not ascribed blame as yet but I'll go ahead and conjecture it was pilot error, pure and simple.
Somewhere else you mentioned how much "scatter" you thought there was in the debris track and whether to attribute that to the rear-engine placement. I'd venture from watching the aerial photos on the news last night it was purely momentum from the velocity and would have been about the same positions even if t'other way 'round in this case.
Again, back to the earlier point -- we don't have the comparative data of the replicated experiment the other way and I seriously doubt anybody will go to the trouble to model in detail what the results would have been if it were to have been in the other configuration--it simply isn't relevant to anything in the instance so there won't be effort made/money spent.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.