How are the speeds matched of the locos in multiple unit operation?
railfan50036How are the speeds matched of the locos in multiple unit operation?
Electric motors are not geared to each other. Each traction motor is geared to its own wheel set. The motor will move the wheel set as fast as the power fed to the traction motor and the load placed upon the wheel set will permit. If a traction motor/wheel set has more power than load - it will slip - thus locomotives that have electric motor have some form of wheel slip control to prevent continued wheel slippage.
There are hundreds if not thousands of books that can detail every facet of this means of electrical traction.
Never too old to have a happy childhood!
At each notch, a locomotive puts out the same percentage of its total output as the others do, so they all go approximately the same speed.
NorthWest At each notch, a locomotive puts out the same percentage of its total output as the others do, so they all go approximately the same speed.
Yes-freight locomotives all tend to have very similar gear ratios unless they are for specific special jobs. UP tried to equip some "Fast Forties" with faster ratios but gave up as it was not flexible enough.
Back in the 50sand 60s SP had passenger engines with several different gear ratios. The E-units and PAs were geared a little above the FP7s, which in turn were geared higher than the boiler-equipped SD7/9s and GP9s. FP7 and F7 units were tryed with compromise 61:16 gearing before settling on 60:17. By the late 1960s all of the remaining non-commute passenger power including the new SDP45s had been regeared to 60:17, which was supposed to be good for 78 MPH, slightly above SP's 75MPH limit on most lines. Even the SD9E pair rebuilt in 1974 for "reserve passenger service" got 60:17 gears. (The trailing unit in last month's SP office car story was one of those). There were supposedly rules on which units could be mixed. In practice it worked out that the E's and PAs stayed together, the GPs and SDs worked alone, and the FP7/F7Bs stayed together. After 1968 the SDP45s often partnered with FP7s and F7Bs.
Having noted that, SP often mixed other units in, usually with 65MPH 62:15 gears. SSW 9389 in the office car story is one of them. I rode the Zephyr over Donner once with 9389 (in red and gray) leading a pair of Amtrak SDP40Fs as far as Sparks.
Towards the end of their service lives, Amtrak leased out some F40PH's to UP and other freight carriers. They wound up being restricted to intermodal service because of their high-speed gearing.
nfotis NorthWest At each notch, a locomotive puts out the same percentage of its total output as the others do, so they all go approximately the same speed. That works as long as these locomotives have a similar speed scale, with each notch position corresponding approximately to the same speed as other types. Things start to become 'interesting' if you mix freight and passenger locomotives in the consist, as you can imagine... N.F.
Problems with locomotive that have different gearing manifest themselves in two fashions. A 'passenger' locomotive being used in freight service has a higher 'minimum continuous speed' - the speed at which the maximum amperage can be supplied to the traction motor and not have the traction motor overheat and damage the electrical components of the traction motor. There are 'short time ratings' which allow more amperage to go to the traction motors for a defined amount of time. These kinds of restrictions apply to DC traction, not AC traction.
The other side of the gearing issue is that a 'passenger' locomotive can operate at speeds that will have 'freight' geared locomotive suffering mechanical damage to their traction motors account excessive rotational speeds. In current freight and passenger locomotives the traction motors are physically geared to their own wheelset and there is no 'neutral' - if a wheel turns a traction motor turns.
Back in the day, passenger locomotives could be geared for operation at speeds as high as 118 MPH. Freight locomotives have mostly been geared to permit a maximum speed of 70 MPH - on my former carrier GE freight locomotives had a maximum speed of 75 MPH and EMD freight locomtives had a maximum speed of 70 MPH. Timetables limited maximum freight train speeds to 70 MPH for 'premium' intermodal trains with all other trains having a maximum speed of 60 MPH and less.
Balt has it right. Gear ratio only is an issue at lower and higher speeds. In between it just manifests itself by the voltage and amperage for a given speed.
-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/)
One big difference would be older DC electrics where different classes would be incapable of MU'ing. Examples were the GE boxcabs and Little Joes on the Milwaukee and the boxcabs vs 201-202 on the BA&P.
railfan50036 How are the speeds matched of the locos in multiple unit operation?
.
Chicago Rapid Transit, the precursor to today's CTA, ran mixed consists from various eras and body types (40 or 50MPH top speed, wood and steel bodied) in trains of up to eight cars, with plenty of tugging and shoving. Rube Goldberg adaptations of mismatched control systems didn't help a whole lot.
CRT tenant Chicago Aurora & Elgin leased and later bought wooden cars from the North Shore Line that would trainline with their own, but the speed mismatch between the series ended the practice of mixing them pretty quickly.
BigJim railfan50036 How are the speeds matched of the locos in multiple unit operation? Boy, you guys make things so complicated!The best analogy that I have ever seen was to think of this as a Tug-of-War. On one side of the rope we have the train, on the other side of the rope we have the engines.
Boy, you guys make things so complicated!The best analogy that I have ever seen was to think of this as a Tug-of-War. On one side of the rope we have the train, on the other side of the rope we have the engines.
That's cute, butit's already the wrong analogy. We're concerned with what's happening between the MUed units, not behind the drawbar of the last one (except in a more incidental sense)
When one person pulls on the rope with all of their might, nothing moves. When another person grabs ahold and pulls with all of their might, nothing moves, when another grabs the rope and starts pulling, things begin to move and so on and so forth. Everyone pulls the best that they can and the train moves. Eezy-peezy!
But the situation here is more complicated. What if your first person is pulling away and the second person yanks the rope harder, or pulls with more acceleration on the rope and bangs into the first person? Or when the first person takes his time 'warming up' to pull hard, while the second person goes right to it and yanks? Those are the kinds of analogy that apply in this situation, and you are not discussing them.
You could also include (appropriately) some considerations like maximum pull coming only above a certain rope speed, or time at full pull being limited by endurance or excessive temperature, or leverage producing greater slipping or differential force when different 'notches' of exertion are selected by a simple system (or when the analogue of 'making transition' factors into pulling on the rope)
The analogy itself is a reasonable one (if you include some method of incorporating active buff and inertia into the tug-of-war framework). But it is far from 'easy-peasy' dynamically, and you as one of the more experienced engineers on this forum might have far better given us some of your experiences with mismatched consists...
[/quote]
There is / was an A&SAB alcoa road switcher on display in Panama City, Fl that had just a 19 (?) pin MU receptacle not the standard 27 one.
Based on Jerry Pinkepank's article, "Lash 'Em Up", in TRAINS some years back, 27-pin MU cables were the standard with a two-cable arrangement of 12 and 21 pin cables being a major alternate. It also mentioned that not every pin carried a function. There were also some other variants.
Until about the mid-1960's, locomotives stayed almost entirely on their home roads so MU incompatibility was not a major issue.
RME BigJim railfan50036 How are the speeds matched of the locos in multiple unit operation? Boy, you guys make things so complicated!The best analogy that I have ever seen was to think of this as a Tug-of-War. On one side of the rope we have the train, on the other side of the rope we have the engines. When one person pulls on the rope with all of their might, nothing moves. When another person grabs ahold and pulls with all of their might, nothing moves, when another grabs the rope and starts pulling, things begin to move and so on and so forth. Everyone pulls the best that they can and the train moves. Eezy-peezy! You could also include (appropriately) some considerations like maximum pull coming only above a certain rope speed, or time at full pull being limited by endurance or excessive temperature, or leverage producing greater slipping or differential force when different 'notches' of exertion are selected by a simple system (or when the analogue of 'making transition' factors into pulling on the rope) The analogy itself is a reasonable one (if you include some method of incorporating active buff and inertia into the tug-of-war framework). But it is far from 'easy-peasy' dynamically, and you as one of the more experienced engineers on this forum might have far better given us some of your experiences with mismatched consists...
BigJim
Like a reality show, you are creating much more drama than actually exists. None of that has anything to do with "matching" the speeds of the locomotives!
BigJimAnd I rest my case! Like a reality show, you are creating much more drama than actually exists. None of that has anything to do with "matching" the speeds of the locomotives!
I think part of this is that we're using very different definitions of what he means by 'matching the speeds'.
Rather obviously, railroads make some attempt to keep locomotives with similar loading response and gear ratio in consists ... this avoids the very real issues I was discussing.
I had thought this was clear, but in retrospect I see it was an assumption: there is no real attempt to "match" locomotive speeds with any kind of adjustment once the consist is together and on a train. (I have seen some articles about tinkering with things like transition relays when those 'popped' in inconvenient ways, this in a sense and indirectly being related to gear ratio, but I doubt these involved actual "engineering" or perhaps even management- or craft-sanctioned activities.)
One thing to remember, though, is that even in steady-state the loading that locomotives take up based on MU control may not be the 'pull-on-the-rope' sum of all the nominal TEs that, for example, applies to steam multiple-headers, although (as Big Jim notes) the difference is usually not of meaningful importance in running trains. If you look at comparable points on the constant-power graph for locomotives with different final-drive ratios (leaving out conditions of low-power motor rating or high-speed birdsnesting risk) you can easily see the difference in effective drawbar pull at speed dictated by the digital -- and yes, it is digital -- 'run' setting of the MU control. And in essence this is the situation you'd have if you had a steam doubleheader ... with a common throttle and reverse setting for both locomotives that couldn't be 'tweaked'. (That of course is much more extreme than the situation for diesels, but I think it establishes the nature of the "issue".)
I gotta agree with BigJim that everyone is getting carried away with all kinds of details. We need to hear from Railfan exactly what he meant by his question. I assumed he was asking about how throttle setting was passed from one unit to another.
_____________
"A stranger's just a friend you ain't met yet." --- Dave Gardner
Paul of Covington I assumed he was asking about how throttle setting was passed from one unit to another.
But that's defined by the MU settings for the engine governor, and has nothing whatsoever to do with motor gearing (or transition or much of anything else of relevance to the issue of "matching" units).
Since there is evidently little interest in ways of achieving better 'fine matching' of locomotive unit performance, here is one reference that answers Paul's assumed question.
One example of benefit from an enhanced control interpretation is David Cook's patent application 20140365049. There are obvious versions of this approach that could address mismatches in gearing, transition, etc. but we can take them up in a different thread where there is interest.
There is an old saying:Perception means everything, reality means nothing.Folks in my forty+ years of railroading, the reality is, speed matching is a perception...a non-issue...a bunch of bunk! You have GOT to remember, real locomotive axles are NOT geared together in a group like toy trains, so, they don't fight one another. Each axle has its own traction motor, independently geared. They are free wheeling and compliment one another, so they do not care what the other units are doing.
Each unit will only do what is asked of it. AND keep in mind that a throttle notch does not ask for speed, but, that certain horsepower rating that it is designed to give for each particular notch.
BigJim Each unit will only do what is asked of it. AND keep in mind that a throttle notch does not ask for speed, but, that certain horsepower rating that it is designed to give for each particular notch.
Someone correct but we understand that each throttle notch calls for a certain number of Amps ? The amps divided by number of powered axels has that value ggoing to each traction motor. Each motor applies power to axel at that rate. What Amps for each notch is determined by builder and maybe ordering RR. RR engineers look at Amp meter to judge power although only on controlling loco. Dynamic braking also is set by throttle notch and indicated amps.
And your "Amps" calculate to Hp. Depending on the manufacturer, the display will show either or both depending where you look and can display Hp/axle if you so desire.
If I recall correctly, a given throttle notch sets a certain speed for the prime mover (rpm, not mph). It is up to the electrical cabinet (Lemp control back in the DC traction generator days) to provide the appropriate loading for the given prime mover speed. In the case of the Lemp control scheme, the current varied inversely with the voltage giving effectively a constant power no matter what the traction motor speed (at least within limits).
The upshot is, as you said, that the locomotives produce whatever tractive effort they can for a given throttle position and the eclectric transimssion takes care of the rest.
blue streak 1Someone correct but we understand that each throttle notch calls for a certain number of Amps?
There are some good references on the 'technical' implementation of what each notch (or, on some GEs, the intermediate notches) actually do. The basic thing the notches control on older diesels is the 'fuel rack' on the engine governor, which controls the amount of fuel injected to the diesel engine and, somewhat incidentally, its rotational speed. If there were no load applied to the engine crankshaft, even relatively little additional fuel injection will make the engine promptly overspeed to the point the governor limiter kicks in, at which point it will not be making very much "horsepower" at all. So there is also some modulation of the imposed load, which on a diesel-electric is of course done by controlling the power from the traction generator/alternator.
One of the great innovations of digital 'notch' MU is that it gives the preselected combinations of fuel feed and electrical excitation that correspond to fixed (and calibratable) positions. Earlier control strategies allowed separate control over engine fuel feed and generator excitation: this could result in a number of unfortunate things, including 'lugging' the engine (too little fuel with too much load) or producing wild overspeed trip if the field dropped out due to something like ground fault due to banging the carbon out of the brush holders going over a grade crossing. In the days when relay logic was the only real robust technology that could survive in diesel locomotive service, it made sense to have a few 'switchable' settings that could be selected by a simple digital (2^3) code determined by power on three wires (it isn't positional 'binary' code like SCSI switch addressing, but could have been) and that code is what determines the fuel and corresponding amperage delivery for each notch.
The amps divided by number of powered axles has that value going to each traction motor.
Much more to it than that -- consider transition (or the effect of back transition)and the reasons for controlling it. If this were as easy-peasy as you guys are trying to say it is, there'd be no need for fancy Amplidynes or Woodward governors and stuff, and anybody who could read the modified Ward-Leonard patents could design high-horsepower diesels.
The issue with 'mismatch', again, doesn't concern anything as obvious as steady-state power output -- it has to do, for example, with how quickly the locomotive accelerates the speed of the diesel engine when the rack is advanced (feed too much fuel on an Alco and you get more smoke than horsepower until the turbo bumbles up to speed; match power ramp-up to unburnt-fuel pollution on some GEs and it takes 30 seconds, perhaps more, to spin the engine up and control load to get clean exhaust) or the transition timing ... usually related to amperage, but coordinated more or less directly with road speed measured in terms of motor rotational speed and corresponding back EMF ... and as noted this produces jostling or even outright banging (if the draft gear has play) as some units accelerate quicker, but others may accelerate harder once they load up.
Sure, they all wind up pulling on the train, and if your interest is just getting the train over the road (don't ask me to argue that isn't the most important point of the exercise, either!) you'll tolerate any amount of internal twitching that doesn't culminate in wheelspin or microslipping that causes unnecessary or strange wheel wear or misprofiling. People who are concerned with cost-effective motive power maintenance, on the other hand, have additional concerns, and 'mismatched' power in consists can be the source of no few of them.
RR engineers look at Amp meter to judge power although only on controlling loco.
One of the fun things on modern power is that you can look at the ammeter data for other locomotives in the consist, including DPU, and if necessary modulate the power individually by axle (or even cut axles in and out). If you have an AC locomotive with individual inverters, it does not matter if the motors have different gear ratios or wheel diameters (that is emphatically NOT the case for locomotives that skimped by having only one inverter per truck, but I won't digress into why); they can be individually accommodated by tweaking the rate, phase, or effective waveform of the synthesized AC so that all the axles pull their fair share all the time (and can automatically recover from slips almost immediately, and should never slip 'uncontrolled' for more than the kind of durations found in creep control operation).
Of course, on an AC drive locomotive, the speed is determined directly by the inverter output frequency; there is no 'balancing speed' based on electrical output of a generator turning faster or being more powerfully driven, and no need for series-parallel transition. Matching these things is a matter of controlling the inverter firing ... not something that can, or should, be under an engineer's direct control. Then an issue becomes keeping the 'notch' response consistent with older DC units... more control fun, complexity, potential places for problems to crop up or start propagating...
blue streak 1Someone correct but we understand that each throttle notch calls for a certain number of Amps ?
In Run 8 the engine is supposed to be producing its full power, which (usually, at least) means full RPMs on the prime mover. In Run 5, maybe 50% of full power.
Our community is FREE to join. To participate you must either login or register for an account.