Ed (gmpullman) just posted a photo in the Show Me Something thread of a Pennsy T-1 leading some E7s at the head of a long consist. During the Transisiton Era this must have been quite a sight. Questions: In a lash-up with both types of power, which would be responsible to track the speed of the other? I would think it would be easier for the diesels to do the matching.
And, while we're at it, would the lead loco always be the one that would drop off once the grade had been scaled? And if the lash-up ran the entire route, would it have mattered which type of power was assigned as the lead?
Come to think of it, speed matching for multiple steamers must have been quite an art, either where all were on the head end or a pusher was involved. Ah, nostalgia!
John
John, I only have a few minutes right now, but being a WESTERN MARYLAND and B&O fan, I have learned a bit about steam/diesel doubleheading. They both did a lot of it in the 50's.
Basically, it was all done by the seat of your paints. The lead engineer set the pace, the other, or others adjusted by feel and sound.
Whistle/horn signals were used before radios came on the scene, and it was a "learned" skill.
As for how/where they were added, or how/when they dropped off, lots of different answers there.
I will try to share more later.
Sheldon
I had it explained to me once and that was to imagine a tug-of-war.
It really didn't matter how big or how strong each person tugging was — it all added up to overcome the force on the rope.
"Generally" the helper was coupled in front of the road power. That engineer was responsible for watching signals and maintaining proper track speed.
As Sheldon mentions, it was a seat-of-the-pants, i.e. experience, that told the crews how much work was being done. Some railroads favored double-heading or adding a helper to the head end.
Notably, the Pennsy at Horseshoe curve would add a "Snapper" to the rear of the train but often, a passenger train would have the helper tied on to the front.
PRR_Bennington by Edmund, on Flickr
Remember, these people were doing this day-in, day-out for years. After a while, the road foreman, power dispatchers and crews figured out what worked and what didn't.
There was a pretty steep grade on the B&O branch that went past near my house. Usually coal drags would have two, big EM-1s on the head end and another pushing.
One time, weather the crew was trying to take a short-cut or just trying something different, they put all 3 EM-1s on the head-end.
That worked for about ten minutes. The whole draft gear was ripped out of one of the hopper cars near the head-end.
That experiment wasn't repeated.
Today, the modern railroads are using "distributed power" to try to balance the tonnage and the strain on the train/drawbars.
Good Luck, Ed
I read an article, I think in Trains, about helper service and basically the helper engines, once the train started to move, just put their loco in Run 8 and let the headend decide how fast to go after that.
I don't know if tbis is even responsive to the question asked, but it is my general understanding that diesel electric locomotives, and in particular the traction motors, could be harmed by being pushed or pulled faster than they were geared for or intended for. So a GP geared for 60 mph MU'd with an E unit geared for 90 would go 60, not 90.
I do know that on most railroads if a diesel electric is "dead in tow" as part of the freight train that there are speed limits on that train. I once saw a BN freight go surprisingly slowly through Rochelle Illinois and assumed the train was following another too closely and was getting a lot of yellow block signals, but as the train passed midway through the consist was an elderly SW type switcher (not BN). What I do not know is whether that speed restriction is due to concerns about the traction motors being turned too fast, or that switcher-style trucks on SW switchers are not built for speed.
As for steam, the concern would not be damaging the locomotive but damaging the track. It would be impossible I suspect for any diesel electric to pull or shove a PRR T1 too fast, since they were counterbalanced for 100+ mph, but you likely would not want to go much over 40 or perhaps 50 mph for a Class H 2-8-0. The pounding could damage the track.
Dave Nelson
AttuvianAnd, while we're at it, would the lead loco always be the one that would drop off once the grade have been scaled? And if the lash-up ran the entire route, would it have mattered which type of power was assigned as the lead?
For a helper - an engine assigned to help a train up a steep grade - the helper would either usually go ahead of the train's assigned engine, or sometimes on the rear. There were some practical rules regarding pushing on the rear of the caboose; often the caboose would have to be behind the helper.
For a true doubleheader - the train goes the whole way, start to finish, with two assigned engines - it apparently was normal with steam engines that the lighter engine was put in the front. So if a 2-8-0 and a 2-10-2 were doubleheading, normally the 2-8-0 would be in the lead.
dknelsonI don't know if tbis is even responsive to the question asked, but it is my general understanding that diesel electric locomotives, and in particular the traction motors, could be harmed by being pushed or pulled faster than they were geared for or intended for. So a GP geared for 60 mph MU'd with an E unit geared for 90 would go 60, not 90.
Two different things going on here. Diesels have a maximum rpm speed that the armature on the motor can withstand before the armature becomes unwound. Typically not a issue. They also have an overspeed protection that limits how fast the engine can run.
The other issue that is more common is that the 90 mph geared engine draws very high current running at slow speeds. So if the train is pulling hard at low speeds, the 90 mph traction motors will overheat way before the 60 mph engine motors will. Its even worse with an AC engine. They can operate safely at very low speeds (single digit speeds) that would fry conventional SD40 and earlier engines.
Since trains spend waaaaaaay more time running below 25 mph than they do above 60, the problem with mixing different geared engines isn't on the high side, its on the low end of the speed curve.
dknelsonWhat I do not know is whether that speed restriction is due to concerns about the traction motors being turned too fast, or that switcher-style trucks on SW switchers are not built for speed.
In many cases the AAR type B switcher truck is limited to 45 mph because of the suspension (or lack thereof). If the engine doesn't have roller bearings its even lower.
Dave H. Painted side goes up. My website : wnbranch.com
Great info from everyone so far. Here is some more of what I know.
Cabooses generally needed to be all steel to be pushed by a helper.
Most passenger trains needing helpers would have them added to the front. Even with tightloc couplers, it was an issue of passenger comfort.
The following is a description of a typical 70 to 100 car train leaving Baltimore for Cincinnati.
Typically two Mikados would leave Baltimore with 4,000 to 5,000 tons in tow headed for Brunswick, MD. The rulling grade is about 1%, the two Mikes can easily handle train.
At Brunswick, the two Mikes would be replaced with two EM-1's (2-8-8-4's) or two EL class locos (2-8-8-0). Because the next leg of the trip has grades over 2%, double grade, needs double the power.
And as described by Ed, another EM or EL class would be added to the rear for the steepest grades.
And, following up on Ed's comment about all three locos on front pulling a drawbar out, it works like this - as the grade increases the "pull" of gravity on the train increases the load on the drawbars. The same increase in "pull" that requires more power.
The goal with pushers is for the rear engine to push as many of the cars as it can, allowing the locos in front to pull the rest. Steam or diesel, the engineers can feel this. The train is heavier than the rear loco can push, so he just pushes, sort of has hard as he can. The front locos can't pull the whole train, they pull the front half, more or less.
Some of the first ABBA diesel sets on the B&O replaced those helpers described above. They had more power and did not need to be turned for the return push. But it was years before they replaced all those steam locos with diesels for the primary power. So steam and diesel worked together for many years.
Remember, most of this happened a pretty slow speeds, 25-35 mph for the most part.
The B&O used a lot of Mikados as passenger helpers in front of Pacifics or diesels when needed.
Hope this gives you some more insight into this, more later when I have time.
There are several famous photos of practically brand new Santa Fe passenger train consists, with brand new red/silver warbonnet diesels on the point, on Raton Pass, NM, and Cajon Pass, CA, with ex-Norfolk and Western WWI vintage USRA 2-8-8-2 (N&W Class Y-3) helpers on the point. It happened multiple times with multiple passenger diesel models from PA's to F's to E units, also with the big 4-8-4 steamers.
Santa Fe was a very high speed railroad at the time, at least as compared to others, and the old WWI-era compound articulateds were fast enough going UP the hill, but were too slow running light DOWN the hill to get out of the way of the flow of traffic, so they were re-sold to Virginian in 1947, after postwar traffic had abated (one of the eight had been wrecked and was not sold to Virginian but scrapped).
They were Santa Fe road number 1790 through 1797.
PRR8259 There are several famous photos of practically brand new Santa Fe passenger train consists, with brand new red/silver warbonnet diesels on the point, on Raton Pass, NM, and Cajon Pass, CA, with ex-Norfolk and Western WWI vintage USRA 2-8-8-2 (N&W Class Y-3) helpers on the point. It happened multiple times with multiple passenger diesel models from PA's to F's to E units, also with the big 4-8-4 steamers. Santa Fe was a very high speed railroad at the time, at least as compared to others, and the old WWI-era compound articulateds were fast enough going UP the hill, but were too slow running light DOWN the hill to get out of the way of the flow of traffic, so they were re-sold to Virginian in 1947, after postwar traffic had abated (one of the eight had been wrecked and was not sold to Virginian but scrapped). They were Santa Fe road number 1790 through 1797. John
New Jersey International, which at that time was an importer of brass locomotives, published a soft cover book on USRA based 2-8-8-2s and variants, of which the various N&W Y class engines were. WWII traffic levels caused the ATSF to acquire the 8 examples of the N&W Y-3 class for pusher service on Raton Pass. The book makes the point that the Santa Fe had never been a big fan of articulated locomotives anyway, and the Y-3s as you say were slow due to small drivers. Among the small changes they made: removed the brakeman's doghouse on the tender and installed the traditional ATSF number plate on the smokebox front.
The Santa Fe used them for just 3 years but kept them for five. Why? The book explains that the Santa Fe was planning a super version of a 4-8-4 based on using the huge Y-3 boiler on modifications of the frame/chassis of the 3751 class of 4-8-4. Now that would have been some engine! The book includes an artist's rendering of what it might have looked like. Huge sand dome.
Thank You.
Just as a further way to look at this:
Ignore everything you know about model locomotives and how they operate.
Ignore everything you think you know about diesel-electric locomotive throttles controlling locomotive speed (rather than just engine power via fuel rate).
The 'tug of war' in a train with all engines leading is essentially between the train resistance (the thing you might calculate via the Davis formula, involved in calculation of 'compensated grade') and the tractive effort on the rear drawbar of the motive power. The speed is entirely determined where these BALANCE at a given speed under a given momentary set of conditions. It is not something where you twist a dial or pull a lever to get to x mph.
Many people overthink doubleheading to death, worrying about how the 'horsepower is divided'. Doesn't matter except for convenience. Think about it this way: the entire load is on the drawbar of the rear engine. If that engine can't develop enough TE to move the load, the combination will simply be slowing down (until you reach a lower balancing speed) -- this is why single switchers can move enormous cuts of loaded cars at very low speed. But if you have 'something else' pulling on the front drawbar (with the engine making some particular power) this counteracts some of the train resistance (opposing forces balance, remember your Newton?) and so the balancing speed will tend to move up again.
So in practice good crews would know the operating characteristics of their power and run the train accordingly: keeping the engine loaded lower when out of the traction-motor characteristic, not overspeeding the motors in the lowest-geared power, etc. If not in MU the power on the individual engines would presumably be set for best effectiveness.
Now comes complication #1, which is whether the lead engine or the 'road engine' is in control of the power. Normally the brake will be controlled from whatever engine or unit is leading, but when you have a helper or snapper it may be the 'practice' to run the road engine 'normally' as learned from long experience, for performance, and then run the helper as needed to give aggregate 'better performance' (ability to handle higher weight for a helper; getting better speed up a grade for a snapper).
When doubleheading for performance, the situation is a bit different: the 'ideal' would be to run the two locomotives like the two engines in an articulated locomotive. In practice, of course, that isn't what usually happens, but if one engineer is either 'lazy' or carefully counting his incentive beans, you'll see the telltale absence of exhaust smoke from one or the other stack, indicating the other engine is doing more of the work (and using more fuel and water).
With MUed diesel power, the engines all respond to a given 'run' as their engine governors are set, and power goes to the motors as determined by transition setting. There is little that an engineer can do, practically, to influence things like automatic transition, and in the old days of mixed transitions there might be a number of critical speed ranges to 'avoid' lest you get repeated hunting seasons around 31mph as the automatics try going up and down (there is a pun here for the electrically aware). Meanwhile, in the bad old days traction motors didn't take well to being abruptly banged, so power had to be cut 'a few seconds in advance' of things like crossovers or grade crossings to avoid the risk of levitated-carbon flashovers and the like. (Note that I carefully avoid the issue of different loading times for different units and manufacturers, which is a consideration, but one concerning mostly progressive throttling-up time rather than steady-state running).
In the 'model' context, some of this continues to apply, but as the drive usually is very different, sometimes savagely so (particularly if you have a high-pitch worm involved) there may be much more involved in electrical tinkering to get locomotives to run smoothly together. As this is 'prototype information' I'll leave that to y'all.
BTW: I thought many years ago, and I still think today, what a shame it is that ATSF didn't perform the Y3 boiler swap on a locomotive with 3752-style Franklin type B (or, by then, perhaps even type C) rotary-cam poppet-valve gear. That would have been a locomotive to watch! (Although, granted, not nearly as much fun as a N&W A given a Q2 boiler would have been -- especially if it were then also given the lightweight rods...)
Chapter 2:
The original premise of midtrain slaves/helpers is much the same as putting locomotives on the rear: reduce the effect of train resistance on that front drawbar. (That's basically all it does, vs. putting all the power on the front). Pushers to an extent have less mechanical problem counteracting train resistance ... but you run into "problems", first with the need to put really stout structure under cabooses if you care to shove on them, and then with freight cars beginning to pop out of the train if something like PRR's Big Liz or HC1, or a Triplex or the like, gets going too much.
The astute reader will now realize that if there is power pulling on the front, and there is power shoving back in the train, there will be a point where there is zero actual force between a pair of adjacent cars somewhere. This point is called a 'node' and as NDG mentioned it is present in most operating DPU trains -- the node moves, of course, as speed, resistance, and tractive effort from the power change, and you can hear the resulting slack runout or run-in, sometimes very graphically indicating precisely where the moving node is and how fast it is changing. (In my opinion there should be a node somewhere in each section of a consist with midtrain distributed power, but there may be times when nodes shift from one 'end' of a midtrain consist to the other. Remember, the point is to overcome aggregate train resistance with aggregate 'propulsive' effort, and the niceties of where that effort changes from tractive to pushing are more in in train handling, not motive-power allocation.
Overmod....The astute reader will now realize that if there is power pulling on the front, and there is power shoving back in the train, there will be a point where there is zero actual force between a pair of adjacent cars somewhere. This point is called a 'node' and as NDG mentioned it is present in most operating DPU trains -- the node moves, of course, as speed, resistance, and tractive effort from the power change, and you can hear the resulting slack runout or run-in, sometimes very graphically indicating precisely where the moving node is and how fast it is changing....
I saw that some years ago when running a 71 car train on the up and down profile of my DC-controlled layout, using four locomotives.
It was somewhat noticeable with all locos on the head end, but moreso with a pair front and rear, and even greater when the train was split into three roughly equal sections, with a locomotive front and rear, and the other two splitting the train's segments. The run-in and run-out of the slack was almost unnerving, but there were no cars popping out of the train nor any derailments.
It was a fun experiment, but not normal practice, as trains are usually 20 cars or less, with as many locomotives as might be needed, depending more on train weight, rather than length.
Wayne