Railroad Physics

[Highballer]

My brother, who thinks he knows everything, and I were riding a Surfliner that was being pushed from Camarillo to Los Angeles. Since I have more RR experience than he, he asked me, what is the limit to the speed and the number of cars for a train being pushed vs one being pulled. He seemed to think if there were too many cars or if the speed were too high, at some point the cars being pushed would buckle. I love it when I know something he doesn't but the only source for information on the subject that I could think of, is you guys. Don't disappoint me.

 

[sky12065]

My brother, who thinks he knows everything, and I were riding a Surfliner that was being pushed from Camarillo to Los Angeles. Since I have more RR experience than he, he asked me, what is the limit to the speed and the number of cars for a train being pushed vs one being pulled. He seemed to think if there were too many cars or if the speed were too high, at some point the cars being pushed would buckle. I love it when I know something he doesn't but the only source for information on the subject that I could think of, is you guys. Don't disappoint me.

Your brother would obviously be right on the part I highlited above, but I don't think that there is a simple answer to your question. There are too many variable factors involved. Presuming you're talking trains in general and not just Amtrak, factors such as the weight of the consist, terrain of the route and condition of the tracks would have to be factored in to reach that answer.

If you saw The Discovery Channel's recent broad cast of Extreme Trains where it covered the transportation of coal from the mining area in Western PA to the eastern part of PA where the coal gets processed in the making of Electricity, I think that would give you a little perspective of what I'm stating!

 

[the_traveler]

In that episode that Joe mentioned, if you noticed that they had 2 locomotives pulling the "regular" train, and there were 4 helper locomotives pushing the train over the mountains. So you see it's not an easy question to answer.

 

[had8ley]

Some other factors to be considered are; if the brakes are set up on the train; are the cars empty or loaded, the gradient of the track and the type of power being used to push or pull.

 

[Highballer]

I knew I was coming to the right place. Let's keep it simple. All things equal, can you pull more cars than you can push?

 

[access bob]

My brother, who thinks he knows everything, and I were riding a Surfliner that was being pushed from Camarillo to Los Angeles. Since I have more RR experience than he, he asked me, what is the limit to the speed and the number of cars for a train being pushed vs one being pulled. He seemed to think if there were too many cars or if the speed were too high, at some point the cars being pushed would buckle. I love it when I know something he doesn't but the only source for information on the subject that I could think of, is you guys. Don't disappoint me.

for passenger trains. for a number of years and maybe still the MARC double decked trains to DC were the worlds fastest pushed trains. they were scheduled for places at 105MPH. usually in 5 car consists. Not sure what the top speed they were tested by the FRA on the test track but I have been told they can be pushed safely at 125MPH.

now these are specially made cars for this service but I think the track may have as much effect as the number of cars.

now one problem with freight has nothing to do with being push at all but rather the design of the couplers. when a train starts there is a couple inches of slack in the coupler so in effect an engine is starting only one car at a time, while this can be done when pushing it is much harder to control. it is one thing to "jerk" that last car forward it is another to Slam it backwards.

so for passenger trains, I would not suspect there is any problem if done right since passenger trains rarely are more than 10-15 cars long, but on freights that is another story and it would depend based on alll the details listed below.

HOWEVER unlike the MARC (and other push pull trains) controlling a train from the rear would be next to impossible. the "Push-Pull" trainsets have a remote cab in the last car and the engineer can control the train from there, but all the cars used in that kind of train have specialy equipped cars so that signals and control of the engine can be done from the rear.

Bob

 

[Joel N. Weber II]

HOWEVER unlike the MARC (and other push pull trains) controlling a train from the rear would be next to impossible. the "Push-Pull" trainsets have a remote cab in the last car and the engineer can control the train from there, but all the cars used in that kind of train have specialy equipped cars so that signals and control of the engine can be done from the rear.

Don't the freight railroads use a radio control system to operate locomotives not at the front of the train when the locomotives are not all at the front of the train?

 

[Highballer]

This is all good stuff thank you very much. Let me ask the question another way. Would there be different (theoretical) limits to the speed and consist size for the Surfliner going the same way over the same track when it is pushed vs pulled?

 

[the_traveler]

HOWEVER unlike the MARC (and other push pull trains) controlling a train from the rear would be next to impossible. the "Push-Pull" trainsets have a remote cab in the last car and the engineer can control the train from there, but all the cars used in that kind of train have specialy equipped cars so that signals and control of the engine can be done from the rear.

Don't the freight railroads use a radio control system to operate locomotives not at the front of the train when the locomotives are not all at the front of the train?

And freights have an engineer in the cab of the "pulling" loco at the front, and another engineer in the cab of the "pushing" loco at the rear.

 

[wayman]

for passenger trains. for a number of years and maybe still the MARC double decked trains to DC were the worlds fastest pushed trains. they were scheduled for places at 105MPH. usually in 5 car consists. Not sure what the top speed they were tested by the FRA on the test track but I have been told they can be pushed safely at 125MPH.

When I took the NJ Transit Atlantic City Line in the spring, I was surprised to find it was push-to-ACY, pull-back. Makes sense, of course--there's nowhere to turn the train at ACY, and it keeps the engine fumes away from the station at all times. But I'll admit there were stretches where we were up to speed (79? faster?) in push mode where I was mildly terrified, even though I know it's completely safe (cab car up front). For some reason, the Keystone in push mode has never given me that feeling. Just the unfamiliar (to me) track and the initial surprise we were in push mode, I guess?

 

[access bob]

This is all good stuff thank you very much. Let me ask the question another way. Would there be different (theoretical) limits to the speed and consist size for the Surfliner going the same way over the same track when it is pushed vs pulled?

highly unlikely.

since most of the track used is not high speed track anyway I doubt it gets close to it's design limit either way.

Bob

 

[Rafi]

for passenger trains. for a number of years and maybe still the MARC double decked trains to DC were the worlds fastest pushed trains. they were scheduled for places at 105MPH. usually in 5 car consists. Not sure what the top speed they were tested by the FRA on the test track but I have been told they can be pushed safely at 125MPH.

When I took the NJ Transit Atlantic City Line in the spring, I was surprised to find it was push-to-ACY, pull-back. Makes sense, of course--there's nowhere to turn the train at ACY, and it keeps the engine fumes away from the station at all times. But I'll admit there were stretches where we were up to speed (79? faster?) in push mode where I was mildly terrified, even though I know it's completely safe (cab car up front). For some reason, the Keystone in push mode has never given me that feeling. Just the unfamiliar (to me) track and the initial surprise we were in push mode, I guess?

MARC does, in fact, push the all-Kawasaki MARC III bilevel trains with HHP or AEM-7 engines at 125 MPH for certain stretches of the NEC between Perryville and DC. And yes, it is the fastest pushed train in the world as far as I know, and is the fastest commuter train in the country.

The Atlantic City line on NJ Transit can be a little startling simply because the tracks aren't in as good condition as other tracks where you're used to traveling that speed. It's not that they're in bad condition, it's just that they're a little bumpy in places.

Rafi

 

[GG-1]

Aloha

What I can't decide is the ride more comfortable, noticeable, etc between Push and pull operation. If so then there would be an Impact on this discussion.

Aloha

 

[access bob]

Aloha
What I can't decide is the ride more comfortable, noticeable, etc between Push and pull operation. If so then there would be an Impact on this discussion.

Aloha

I ride MARC all the time and I have never noticed a difference.

now when a diesel gets put on one of the MARC trains that usually has an electric. you can really feel the difference, sometimes feel like you want to get out and help get it going. not that the diesel is slow, it is that the electric is that much more powerfull accelerating.

Bob

 

[Joel N. Weber II]

What I can't decide is the ride more comfortable, noticeable, etc between Push and pull operation. If so then there would be an Impact on this discussion.

The two round trips I made from BOS to PVD on MBTA trains in the spring of 2007 left me with the impression that push mode was less comfortable. On the other hand, I didn't notice that effect on the Fitchburg Line trip the OTOL group went on 6 days ago, nor have I really ever noticed it on the many trips I've taken on the Fitchburg Line in past years.

The Fitchburg Line trains are currently single level (and I expect they will be at least until the Rotem coaches are delivered to the MBTA); the coaches the MBTA uses on the Providence Line trains are bi-level. I was probably sitting on the upper level; it's occuring to me now that comparing the upper level to the lower level might be interesting.

I think the several trips on Providence Line and one outbound (pull) trip on the MBTA Franklin Line are the only experience I have riding on bi-level trains.

The Providence Line track is probably in better condition than the Fitchburg Line, given that the Providence Line is the NEC, so whatever I was disliking isn't likely to be a track condition issue.

 

[OlympianHiawatha]

From time to time the Heartland Flyer operates with a P42 on both ends rather than the standard "cabbage" and when in that configuration, I have noticed both engines are running. That makes me wonder if both are providing traction power, or if the rear engine is only in "hotel" mode.

 

[had8ley]

HOWEVER unlike the MARC (and other push pull trains) controlling a train from the rear would be next to impossible. the "Push-Pull" trainsets have a remote cab in the last car and the engineer can control the train from there, but all the cars used in that kind of train have specialy equipped cars so that signals and control of the engine can be done from the rear.

Don't the freight railroads use a radio control system to operate locomotives not at the front of the train when the locomotives are not all at the front of the train?

Yes, most of the time. There still is manual pusher service (manned by a live crew) at certain places (the Huey P.Long bridge in New Orleans is one of those places. There is also distributed power placements in the middle of the train in addition to the rear~ they are remote controlled.

 

[PetalumaLoco]

Here's the smart a$$ answer; one. And that car pushes the next. And so on.

 

[jis]

MARC does, in fact, push the all-Kawasaki MARC III bilevel trains with HHP or AEM-7 engines at 125 MPH for certain stretches of the NEC between Perryville and DC. And yes, it is the fastest pushed train in the world as far as I know, and is the fastest commuter train in the country.

Amtrak Keystones also operate at 125mph on the NEC in push (as well as pull) mode powered by AEM-7s and using a converted Metroliner car as cab car. But much before any of this happened all ECML London Kings Cross - Edinburgh Waverley trains have operated in push mode while traveling in one direction (I forget which now) at 125 mph, with a DVT in the front and a Class 91 pushing. And they are by far much longer trains than either the MARC ones or the Amtrak Keystone ones.

 

[Guest_Yerry_*]

Passenger trains run in push mode with a max speed less than forwards simply because there's not much test data at all on pushing at high speeds. AAR isn't testing because there's no perceived need (or ability) to increase pushing speed.

Santa Fe had all kinds of problems with "stringlining" on the old layout of Cajon Pass; In the '60's, high center-of-gravity, 86' cars would tip over INWARDS on curves when the locos PULLED too hard.

KCS once had issues with long trains with helpers when loaded grain hoppers (again, high COG) would be PUSHED off on the OUTSIDE of a curve; this led to the realization (and their removal) that superelevation and long, heavy trains don't mix. SP was very quick to remove superelevation almost immediately afterwards.

But Amtrak cars have a much lower COG, due to the enormous amount of steel in the wheels, trucks, center sill and undercarriage equipment (AC, water, batteries). Even a Superliner car has a much lower COG than one would suspect: the heaviest equipment was strategically located to ensure this. Their highest COG car is an old Heritage baggage-- which is so high, comparatively speaking, that during the SDP40F debacle, putting a lower COG deadhead coach between the loco and baggage car changed tracking characteristics so much that it was used frequently in an attempt to solve the problem. SDP40F + high-COG car + reverse curves with no tangent in between = wreck.

When the Superliners were built, they were severely tested on the AAR test track in Pueblo, CO, at 120 MPH pulling and (WOW) 90 pushing.

There's no slack in a passenger train. Passenger cars use locking couplers (Tite-Lok is a brand name) to both eliminate slack and also to keep the cars together (keeping couplers from behaving like a jousting lance) in case of a wreck. They have the drawback of making "Pogo-ing" very obvious, high-speed combined with hard pulling would actually stretch the entire train slightly (less than one inch in a 100 ft long train), and the train would bounce forwards-and-back one or two times a second. The old F40 was great at this. Pogo isn't a tracking problem, and it disappears on curves anyway.

Finally, every passenger car has a snubber on each truck. That's the nearly-horizontal shock absorber connecting the side of the truck with the side of the car frame. Trucks don't swivel freely on curves; the track forces the truck to turn (or straighten out). In the event of a totally-off-the-rails derailment, the snubber keeps the truck from suddenly turning. Cars with straight, parallel trucks rarely tip over.

Unlike toy trains, real couplers have no sideways forces holding cars together. i researched all the above after a ride where some kid was seriously worried because his trainset couldn't run in reverse at all.

The only change in ride quality I've ever encountered centered around my car being located next to the loco (or not)-- they're heavy (190 tons) and the track rebounds UNDER the adjacent car when pulling, but not when pushing.

 

[Guest_Yerry_*]

Hundred foot train? Oops. Shold be "1050 feet"

 

[abqdave]

I wonder if the answer to how much/how fast pushing vs. pulling isn't the going part, but is the stopping part?

If you had a very long train going very fast and it had to stop very quickly...if it was pushing, would there be a danger the couplings wouldn't hold everything together?

 

[had8ley]

I wonder if the answer to how much/how fast pushing vs. pulling isn't the going part, but is the stopping part?
If you had a very long train going very fast and it had to stop very quickly...if it was pushing, would there be a danger the couplings wouldn't hold everything together?

Your most delicate time in running a train is stretch braking. On freight trains of old there were no dynamic brakes. Dynamics bunch the train up and gather all the slack through reversal of tractive motor effort. This action relieves the stretch action on the car equipment. The couplers (knuckles) are the most usual pieces of equipment to fail when pulling a train. You do not usually experience equipment failure when shoving a train unless you shove too hard and derail. The knuckle is a crescent piece of metal that weighs about 50 pounds or so and is hard to replace if you break one enroute. The most devastating feature of "tearing up a train" is pulling a drawbar that connects the two cars together. If you are traveling south and and pull the drawbar out of the south end of a car you have no way of moving the car unless you have a cable. It can get very interesting especially if you have to go ten miles or so to set out a cabled car, that is deep in the back of your train, in a siding. I was very fortunate and never got a knuckle or drawbar but we had engineers who got one often; especially on the Huey P. Long bridge.

 

[spacecadet]

Unlike toy trains, real couplers have no sideways forces holding cars together. i researched all the above after a ride where some kid was seriously worried because his trainset couldn't run in reverse at all.

I'm trying to get my head around the relationship between model trains and real trains, because I always have the same problem (cars just get pushed off the tracks) and I always thought those same forces must exist on real trains.

Why wouldn't they? If you have a curve with two 85' cars on it, they're going to be at an angle to each other, and the locomotive is going to be at a further angle. Isn't the pushing force of the locomotive on the first car going at least partially in a direction that's off the track? (In other words, the force is being directed in a straight line from the locomotive, but the track is curved.) And with the weight of additional cars down the line, wouldn't that force going in the "wrong" direction at some point be greater than the force available to overcome the inertia in the direction of the track (causing a derailment)? I mean, say you're pushing 50 loaded oil tanker cars, and your train is stopped on a sharp curve. That would seem like it would take a large amount of force to overcome the train's inertia in the direction of the track, but to overcome *one* car's inertia (and the force holding it on the track) in the straight-line direction going *off* the track would seem like it would require considerably less force.

(I would think the same would be true pulling, but somehow pulling seems easier on model trains, at least.)

Or am I just way off base?

 

[had8ley]

Unlike toy trains, real couplers have no sideways forces holding cars together. i researched all the above after a ride where some kid was seriously worried because his trainset couldn't run in reverse at all.

I'm trying to get my head around the relationship between model trains and real trains, because I always have the same problem (cars just get pushed off the tracks) and I always thought those same forces must exist on real trains.

Why wouldn't they? If you have a curve with two 85' cars on it, they're going to be at an angle to each other, and the locomotive is going to be at a further angle. Isn't the pushing force of the locomotive on the first car going at least partially in a direction that's off the track? (In other words, the force is being directed in a straight line from the locomotive, but the track is curved.) And with the weight of additional cars down the line, wouldn't that force going in the "wrong" direction at some point be greater than the force available to overcome the inertia in the direction of the track (causing a derailment)? I mean, say you're pushing 50 loaded oil tanker cars, and your train is stopped on a sharp curve. That would seem like it would take a large amount of force to overcome the train's inertia in the direction of the track, but to overcome *one* car's inertia (and the force holding it on the track) in the straight-line direction going *off* the track would seem like it would require considerably less force.

(I would think the same would be true pulling, but somehow pulling seems easier on model trains, at least.)

Or am I just way off base?

No; if fact you hit a home run. Imagine a rubber band expanded from the front of your train to the rear which is in a curve. Just where does the rubber band go? Straight to the rear in a straight line, not curved. Another factor is if the train brakes are released or applied. You can push all day if the brakes are set up on the rear of the train and all that you will probably accomplish is called a buffed derailment. I have seen cars literally jump off the track, in a sharp curve, when the train brakes were not fully released and excessive motive power was exerted on the leading cars of the train in a forward movement direction. I haven't seen model trains with air brakes yet but perhaps this will give you some insight as to how it is on a freight train.