On 17/05/2013 22:44, Stephen Sprunk wrote:
> On 17-May-13 11:10, Nick Fotis wrote:
>> On 17/05/2013 18:04, Stephen Sprunk wrote:
>>> A modern unit coal train would be ~15,000 short tons.
>>
>> A question is, do you really need such large trains, or (on a
>> double-tracked mainline) shorter trains at 10.000 tonnes each would
>> flow better along the other types of trains?
>
> It's a matter of efficiency. There is a fixed amount of empty track
> between trains for safety reasons. Therefore, to maximize the amount of
> freight moved per track used, you want each train to be as long as possible.
I suppose you mean 'braking distance', right?
If you have a serious signaling system like LZB, it computes for your
train automagically the braking distance (you tell it which of the
pre-defined braking curves to use), and keeps the correct distance from
the preceding train.
I hope/suppose that the coming PTC will offer such functionality.
> For instance, say a 15,000tn coal train is 2mi long, plus there's 2mi of
> empty space behind it. Two such trains consume 8mi of track. Now say a
> train is only 10,000tn and 1.3mi long, plus the same 2mi of empty space
> behind it. You now need three such trains to move the same amount of
> coal, which consume a total of 10mi of track: a 25% increase in track
> usage for the same amount of revenue. Oh, and now you have to pay three
> train crews instead of two, a 50% increase in wages.
You assume a fixed braking distances for all kinds of trains (based on
the worst case), which is a very inefficient use of track capacity.
>> A major capacity constraint on such lines is the speed differential
>> between various classes of trains.
>
> If you make the trains shorter to speed them up, you increase the energy
> costs _and_ the amount of empty space between trains, further reducing
> efficiency (for those trains, at least).
The energy cost can be so low, compared to diesel fuel (thanks also to
regenerative braking and driver advisory systems), that you will be able
to afford different strategies.
> Note that bulk unit trains are the most profitable, while intermodal
> trains barely break even (and passenger trains lose money), which means
> you're messing with exactly the wrong trains.
This happens because the diesel fuel consumption behavior directs you
towards a model of 'the minimum horsepower which can go over the hill'.
With electric traction, this behaves differently, as you can get back
the energy you spent going uphill when going downhill (provided that you
have large enough electrified sections that there will be trains able to
consume that power)
>> Two pairs of these locomotives can haul a 14.500 metric tons train at
>> a 1% grade at a speed of 60 km/h (37.2 mph), and at 93.2 km/h (58
>> mph). Is this good enough for you, or should we add more power per
>> locomotive? :-)
>
> That's probably sufficient. Still, two pairs of IOREs per train? That
> just feels _wrong_.
Well, things you have learned with diesel-electric are not valid anymore
when you go straight electric.
Note that one 2-section 12-axle IORE hauls 8.500 metric tonnes iron ore
train in Sweden.
Chinese pull 10.200 metric tonnes per 2-section 8-axle locomotive, with
"only" 25 metric tonnes/axle.
> Seems like it'd be better to use three pairs of something smaller, which
> would provide more operational flexibility.
No reason to do that. The reliability of electric locomotives, if you do
not do any major gaffe, exceed the one attainable by a diesel
locomotive, thanks to their much simpler construction (much fewer
mechanical elements, mostly solid state electronics and some electric
motors).
I remember somewhere reading that the 125 mph Class 101 of German
railways were guaranteed for up to one train-stopping failure per
million kilometers. And even Indian railways (Southern region) report
than 2.5 failures per million kilometers on their electric locomotives.
>> A very nice thing about electric locomotive is that you do not
>> consume more energy than you really use (no energy is used when
>> idling), so you can add an extra locomotive without affecting your
>> energy consumption much.
>
> OTOH, having that extra loco idling means you're wasting scarce capital
> assets and increasing the maintenance required for no good reason.
The amounts of maintenance needed by straight electric locomotives are
generally very low compared to diesel-electrics.
And when an electric locomotive idles while having the pantograph up, it
is practically the same as parked (only maybe the air compressor operates).
> Furthermore, US freight RRs don't seem to care much about energy
> consumption; they leave trains idling in sidings for hours at a time,
> leave locos idling in yards overnight, etc. And, of course, they
> haven't electrified, which would provide tremendous energy savings.
Since they are (still) passing their cost of fuel to the customer (and
have a fuel surcharge to boot), why should they care (yet)?
> Double-tracking trunk routes, not improvements in traction, is what they
> believe will produce the biggest efficiency gains, as shown by where
> they're investing their capital.
If you trace back my initial comments on this thread, the reason I
suggested BNSF Transcon as the most likely candidate is that they have
double-tracked almost all the route already.
And they have pulled out of Wall St., so they are free from the tyranny
of quarterly reports and dividends.
>> The general practice about using the trailing pantograph is that if
>> you get catenary damage, you protect the front pantograph from
>> damage, so you can move ahead (debris from catenary failure could
>> damage a trailing pantograph in the 'up' position).
>
> Is there no practical limit to how much power can be collected by a
> single pantograph?
The limit is due to amperage (and heating due to resistance). When you
up the voltage to at least 25 kV, you have a large enough margin (you
may prefer to go as high as 50 kV, if the cable clearances are met, and
you can install fewer substations en route)
The largest single locomotives I know are the 9.6 MW 6-axle motors
currently running in China, and the 10 MW dual-section HXD.
The most powerful trainsets are the 16-car Siemens Velaro: 16 MW for the
Eurostar (e320), 19.2 MW for the Chinese 380CL (I think these use two
pantographs, spaced 8+ cars apart):
http://en.wikipedia.org/wiki/Siemens_Velaro
N.F.