Gen-Set Locomotives

UP has recently ordered large batches of gen-set locomotives from National Railway Equipment (for California) and Railpower Technologies (for Texas). The principal of using multiple (2 or 3) truck-size diesel-engine-with-generator sets per locomotive with one set operating full-time and the others fired up only when needed sounds good in principle as a way of reducing emissions. Is there any real-world experience that NRE and Railpower are using as a basis for their designs or is a big leap of faith involved here?

Not in the RR industry. What it sounds like what they are doing is kind of like the power industry and gen sets when the load increases they light off another one. sounds good in principle but look for some problems. I wonder how they mechanics are going to feel when they are changing out parts on a Cat or Cummins engine.

Railpower’s stock to a big fall last week due to warranty concerns. The railroads have been using the Green Goats in service so far heavier than forecast, decreasing the life of the batteries.

http://ca.news.finance.yahoo.com/01022006/2/finance-railpower-stock-falls-firm-takes-23m-28m-provision-q4.html

I hope they can overcome these problems. If you ever watched a set of hump engines or bowl engines you saw a tremendous amount of smoke and smog emitted by them. Having less smoke/smog would be a good thing.

No leap of faith required here at all. The engine and main-generator characteristics are well established, as are most of the maintenance characteristics, MTTF, etc. Aside from the possible economies of scale (and service, maintenance, parts, etc.) in using largely OTS “truck” motors for rail, it becomes much easier to use and control some of the final-stage pollution technologies (such as ammonia injection and particulate traps) with several small engines than with one big one.

I’m tempted to say that you’re seeing the ‘shakeout’ of overly-hyped “green” technologies when applied to real-world railroading, but that would be a bit of a cheap shot. I don;t know the design cycles for ‘switcher’ loading that RailPower used, but they might have assumed shorter cuts of cars, lower speeds, and less ‘kicking’ than actual yard crews use – Ed Blysard, for example, can probably provide very useful insight into what switch crews like to use, and Randy Stahl can translate that into the electrical ‘consequences’ on the motors, and thence to the batteries and controls. Note in particular that very large ‘spikes’ of regeneration power (from the traction motors) may be above the ‘safe’ charging rate of some of the battery cells, and many cycles of repetition of even short overloading may be causing unanticipated cumulative damage.

Historically, IIRC, some (perhaps all?) of the original Clessie Cummins proposals for PRR back in the '20s involved multiple gensets using comparatively small engines, and I suspect the Baldwin modular locomotive (6000) did not succeed mostly for reasons unconnected with the idea of multiple ‘slot-in’ power gensets per se. Of course, neither of these were primarily designed with the explicit intent of trimming power (rather with achieving high power with the limited capacity of individual diesel engines at the time)

One assumption is that maximum speed demanded of the genset locomotives be restricted – I think it’s reasonable to build a class of ‘dedicated’

I agree with overmod. It’s probably just a matter of tweeking the Generating capacity to match the average demand of the load. It always did seem to me that the size of the generators were too small for the duty cycle of a switcher in yard duty, let alone the duty cycle for going on the road to do local jobs. Then again I read somewhere recently that the genset locos were designed in a building block fashion where additional generators could be added to match the duty cycle needed for the service they are in. That makes them fairly inflexabe for use in other services though.

[quote]
QUOTE: Originally posted by Overmod

No leap of faith required here at all. The engine and main-generator characteristics are well established, as are most of the maintenance characteristics, MTTF, etc. Aside from the possible economies of scale (and service, maintenance, parts, etc.) in using largely OTS “truck” motors for rail, it becomes much easier to use and control some of the final-stage pollution technologies (such as ammonia injection and particulate traps) with several small engines than with one big one.

I’m tempted to say that you’re seeing the ‘shakeout’ of overly-hyped “green” technologies when applied to real-world railroading, but that would be a bit of a cheap shot. I don;t know the design cycles for ‘switcher’ loading that RailPower used, but they might have assumed shorter cuts of cars, lower speeds, and less ‘kicking’ than actual yard crews use – Ed Blysard, for example, can probably provide very useful insight into what switch crews like to use, and Randy Stahl can translate that into the electrical ‘consequences’ on the motors, and thence to the batteries and controls. Note in particular that very large ‘spikes’ of regeneration power (from the traction motors) may be above the ‘safe’ charging rate of some of the battery cells, and many cycles of repetition of even short overloading may be causing unanticipated cumulative damage.

Historically, IIRC, some (perhaps all?) of the original Clessie Cummins proposals for PRR back in the '20s involved multiple gensets using comparatively small engines, and I suspect the Baldwin modular locomotive (6000) did not succeed mostly for reasons unconnected with the idea of multiple ‘slot-in’ power gensets per se. Of course, neither of these were primarily designed with the explicit intent of trimming power (rather with achieving high power with the limited capacity of individual diesel engines at the time)

One assumption is that maximum speed demanded of the genset locomotives be restric

Batteries and their control are the biggest problem in all of the hybrid designs, whether they are rail or truck or automotive. For all we take them for granted – and we all do, get in the car and turn the key – they are astonishingly delicate beasts. As has been noted, they are sensitive to overcharging, excess charging rate, too deep a discharge, and excessive discharge rates – any one of those can do them in. For the real old-timers, submarines had the same problem! Having absolutely no factual basis to draw on, I would imagine that one of the big things that RailPower and other folks are working on is the computer controls of charge/discharge rates and amounts to optimize the life of the batteries vs. the cost of running the whole rig.

It will take a while to get it right – but in the meantime, they do seem to be on the right track. As it were… sorry about the pun!

UP’s ordering approach of two batches from different suppliers - with deliveries over a two year period is a sound way to invest in new technology. Of course it helps that part of their purchase cost is offset by local environmental incentives. At the end of this purchase, UP and suppliers will either have most of the bugs worked out or shown the technology to be unworkable in RR switching applications. I’ll bet on the former.

dd

These Gen-set locos sound a bit like the twin engined Class 17 built by Claytons diesels that British Rail ordered in the 1960’s. These had a central cab and a 450hp engine under each bonnet. These locos were intended for heavy switching and short trip freight workings (which were rapidly disappearing with line and yard closures taking place all the time). The idea was that for lighter duties one engine could be shut down.

Alas these locos were not a success. The extra complexity of their systems meant they were more costly to maintain than the older (1957 vintage) 1000hp single cab English Electric class 20’s. As a result BR decided to scrap the Class 17s, in some cases after less than 5 years, and ordered another batch of the Class 20’s. To-day only one class 17 survives, having been sold off to an industrial line and then preserved, whereas Class 20’s are STILL in use with some of the open access operators.

Curiosuly, the problems BR had with the Class 17’s did not deter their builders from a further experiment. This loco, known only as DHP1, had TWO 400hp engines under each bonnet. It undertook a few trials on BR but its hydraulic transmission (its designated DHP1 stood for Diesel Hydraulic Prototype 1) meant BR were not very interested. It wound up as works switcher at Clayton’s plant in Derby.

40s years on though, we now have DMU’s with hydraulic transmission rather than loco hauled passenger trains so perhaps the Gen-set’s time has come. After all the 5 car “Adelante” DMU’s which have a 750hp engine in each car can keep time with one engine out of use, so this gives added resilience. And electronics are a lot more robust now. We shall see.

Advances in computer technology may make control of mutliple engine DE’s a lot more robust, maintenance-free, and practical. But anyone know how GE’s forey into Hybrid technology for road power is doing?

One question I do have is how much of the foray into these new technologies is due to government funding. I know that at Soo Line the Cat engined locomotives were well thought of right up to the time of their first overhaul. They required an overhaul at 4 years versus 8 years for a GP38-2 in the same service. Cost about the same too. I see a continuing decline in the need for switching locomotives and light roadswitchers. The carriers may view these locomotives as a bridge until they can eliminate loose car railroading.

I’m calling this thread back up to generate some more thought. Although gen-sets have been proposed primarily as a way of reducing emissions, could they also stand on their own as a way of reducing fuel costs? I’m primarily looking at regionals, short lines and terminal operations outside of California and Texas.
A shortline or a roadswitch job on a regional or an industry job on a terminal road could use a gen-set locomotive to use just the right amount of horsepower to do the job. If the job is light, only one engine may be running; if the job is much heavier on a different day, two or three engines may be on line. Fuel usage would vary depending on the need.

Is this wishful thinking or could it really work?

All those grain elevator switchers would probably be a good match for a genset locomotive. Locally, a cogen plant uses an old SW to move cars (delivered by CSX) for unloading. The switcher is usually doing one of two things - shuttling cars between the yard and the unloading track, or moving cars up one at a time to be bottom dumped. I’m guessing the SW is around 900 HP (haven’t taken the time to determine the model), and they certainly don’t need all of that to push 10 cars 40 feet at a time…

A grain trans-load facility in Illinois that I used to be familiar with shifted long strings of grain hoppers using just a cat tractor and a chain.

Again, anyone know how the two GE Road-Hybrids are doing on the road? Are they still in use or back at the shop?

MIT has announced (check www.mit.edu) the development of a practical prototype litheum battery that can be developed into a production model. This will hopefully cause GM to rethink its position with respect to electric cars, and strenthen its market share and development of hybrids as well as benefit the other USA automakers. Possibly Toyota and Honda may even use this type of battery to improve their existing popular products. (Toyota has said nothing but hybrids in just a few years.) I think we can all hope for a comeback for GM with really advanced design electric cars and without the danger and lower (overall) efficiency of fuel cells.

But this should really impact possibilities for “green” diesel locomotives.

Missed this thread the first time around…
In flat yard switching service we typically switch out two 100 to 120 car trains a shift flat switching, per yard job, with 3 jobs working at the same time.

If lady luck is smiling on us, there are large “cuts” of cars going to the same track and we can bring them around and just shove in, but the norm is the train being comprised of almost all singles, a few doubles.

So look for the trains being “cut up” into 30 to 40 car cuts, with anywhere between 70 to 90 “pins” or kicks per train switched.

Because we switch down several leads, as opposed to a hump job pushing them over a hill at constant speed, or a hump yard trim job dragging a track out of the bowl and setting it over intact in another track, the flat switching locomotive goes from standing still to notch 6 or 8, then back to a stop for each kick.
Most moves involve a kick, then stretching back past the lead switch, and kicking down the other lead, backing up, stopping, kicking down the next lead, so forth and so on.

So count on the locomotive having to go from a standstill to full bore, and then stopping and reversing direction for just about every car kicked.

On a 90 “pin” train, that’s 180 short, high power requirement moves per train switched, or 360 times per shift for one switch crew, a demanding job even for a “normal” locomotive.

Not being privy to all the technical aspect of the green goats, I would suspect that the batteries were designed to be drained at a fairly constant rate, and charged in like manner, as opposed to the short, repetitive high draw use of a flat switcher.

With the gen set type, I don’t see the locomotive moving at a constant rate of speed for any long period of time, just the short burst and stops we do, and I doubt the generators could keep the batteries charged enough, or recharge them fast enough

Imagine the hammering your automotive battery would take if you starte

In my opinion , the engineering on these machines should have included new low speed traction motors. I’m thinking that something along the lines of a axle wound armature similar tp the Bi Polar design . Batterys do very well at making current but quickly lose voltage , the high speed 4-1 gear ratio motors that are used in current locomotives are not designed for low voltage .
Randy

And, with Randy’s idea, there would be zero lag time between the throttle being opened and the locomotive moving, almost instant motion, no wait for the thing to load up…

Hey, Randy,you should write them a letter!

Ed

Randy, could you not achieve similar results by radically changing the gear ratio on an ‘existing’ nose-suspended design, and keep at least the general frame and armature common with other motors? (You’d need a different gearcase for sure, but wouldn’t that be a relatively small inconvenience… and it should be practical to make the ‘larger’ pinion as an adapter that would fit, and perhaps laterally ‘float’, over the smaller ‘standard’ pinion profile on the armature…