There must be some railroad aficionados on the SDMB who can answer this.
Let’s jump in the Wayback machine and travel back to West Virginia in 1912. Railroad trains are pulled by steam locomotives. There is no radio. Coal trains, pulled by huge and powerful locomotives, regularly traverse this hilly terrain. They travel slowly but haul gigantic tonnages. The trains are heavy enough so that it is common (I think) for one or more locomotives to be attached at the end or mid-train to assist getting the train over grades.
There were no automatic locomotive controls back then. Each locomotive had an engineer (driver, for our friends across the pond,) in the cab.
Question: How did the locomotives match their tractive effort so as not to be a drag on the train nor cause a possible derailment by pushing harder than the locomotive(s) in front of it?
I’m really only interested in those situations where mid-train or rear-following locomotives are present. I presume that multiple locomotives in front somehow equalized their effort just by virtue of being coupled together.
They could also have used wired telegraphs of some sort to signal between them. It’s not all that far from the front of the train to the middle, and the cars are all connected anyway.
While we’re at it, you also want communications with the brakeman in the caboose.
I’ve not heard of wired telegraphs being used. It would require every train car to have the telegraph wire installed, with connectors on both ends. Doesn’t sound practical.
My understanding is that communication was mainly by whistle signals.
Also, for a double-header (two locomotives in front), the engineer of the 2nd locomotive could watch the brake line pressure to know when the lead engineer applied the brake. So he’d know to back off on the power. For a pusher at the back of the train, I think the brake line was not usually connected, so this wouldn’t be an option.
The tractive effort doesn’t need to be matched. At one extreme you have one locomotive pushing the whole train; on the other extreme you have one locomotive pulling the whole train. Either extreme, or anything in between, would work OK if the total power was adequate. The problem you want to avoid is having the lead engineer apply the brake (which would apply the brake on the whole train, but not all at once - there is a delay as the pressure drop propagates to the back of the train) while the other locomotive was still applying a lot of power. If that happens, the rear locomotive would be pushing cars into the cars with brakes applied, so that could result in derailment.
Could they tell if they were doing more pushing vs pulling based on the cars taking slack in or out of the couplers? There’s maybe an inch or so of play, right? So if a pushing locomotive applied more force than the pulling locomotive in front you’d get a bang bang bang as the cars in between were pushed together, same if the pusher then let up.
Yes, couplers have slack. I suspect what ends up happening is, the front cars are pulled by the front locomotive (i.e. couplers in between are stretched), the rear cars are pushed by the rear locomotive (couplers compressed), and the cars in the middle jostle back and forth as the balance changes.
If the coupler at the front of the pusher locomotive is stretched, it means the pusher locomotive isn’t pushing at all. Instead the whole train ,including the rear locomotive, is pulled by the front locomotive.
And the Midland Railway followed soon after, using a cable-and-pulley system (something I only found out today, so I’m quite keen on sharing this).
However I’m sure that none of these systems were used when double-heading, or banking, a heavy train with steam locomotives. The old-time railways relied entirely on the skill of the drivers.
Here is a photo by the remarkable Ben Brooksbank showing four (4!) banking locomotives pushing a train up the Lickey Bank in the UK. These locos were not coupled together, and were not connected in any way; but they all worked as a team to push a heavy oil train up the hill.
Interesting, and surprising to me. I had always thought one of the many advantages of diesel-electric locomotives was that they could be electrically synced and so it was easy to have multiple locomotives, so much longer trains became possible, and indeed it’s very common to see multiple diesel locomotives. Until reading this I would have assumed that running multiple steam locomotives wasn’t even possible, or at least not practical. Intuitively one tends to think that the power settings would need to be very closely synchronized, but apparently not.
But as long as they’re all pushing, not pulling, there’s no need to connect them at all. If any loco starts falling behind the one ahead, that’s the operator’s cue to add more power.
I may have understated the difficulty in my earlier post. Wikipedia entry on double-heading talks about skilled crew being necessary, and derailment risks when locomotive controls aren’t coordinated properly. (Though that’s slightly different from having a pusher locomotive in the back.)
They weren’t double engines: rather, they allowed the engineer to be at the opposite end of the train, using an almost standard engine, but not turning it around at each end of the trip.
And the Milwaukee line, which used double headers because of the difficulty of their route, was also the developer of an enormous electrification project – because using double headers is so difficult and inefficient with steam.
Lickey Bank is the steepest incline on a British main line. Extra traction was always needed to push loaded trains up it and two engines were specially built for the job. You should note that the pushing engine was not coupled to the train as it would just let go at the top and work back down for the next train.
MT SWAG and a little knowledge.
These were direct drive steam engines. As the engines start lead engine singles start and both engineers open the throttles. And they will increase throttles together. It takes experience and working together . While traveling a small part of the load will transfer back and forth. With out throttle adjustments as one engine takes more of the load it will slow slightly and the other will speed up. The load will rebalance. Maybe one or two of the cars couplings will have the slack taken up or let out. I would not expect this to be a large bank though. As the lead engine crests the hill he will signal the trail engine. Lead engine will begin to close the throttle. and according to speed he will begin to close the throttle. Until he will begin to have to apply brakes. Following engine will be monitoring his speed until it becomes necessary to close his throttle and begin to apply breaks.
As he tops the crest he will signal lead engine.
I guess the fact that they built insanely huge steam locomotives like the Big Boy should tell us how much they wanted to avoid double-headed trains. The Big Boy (including the tender) weighs more than a fully loaded Boeing 747. The largest modern diesel-electric locomotives are half that size because it’s easy to couple a few of those together and have one driver operate them as if they were one locomotive.
p.s. Here’s a neat footage of a preserved steam locomotive assisting a modern freight train up a hill.
And less relevant, but here is a really nice footage of a UP Challenger (one of the largest locomotives ever built) pulling a modern freight train of 143 freight cars all by itself.
Thank you dopers. I can see how whistles would provide adequate communication. But the real skill appears to lie with the engineers. No magic technology, just experience. And evidently two locomotives up front was not as trivial as I had imagined.
If one of the two locomotives is twice as powerful as the other – so what? Why would its pushing twice as hard cause problems?
If the lead engine is good for 60000 lb of pull, and the mid-train engine is good for 100000 lb, then you might as well position the mid-train engine so it can exert 50000 lb of push and 50000 lb of pull. But railroads often (usually?) didn’t bother with that. B&O hauled its coal up 2% Cranberry Grade with one 2-8-8-0 on the front and two on the rear; apparently they managed well enough.
