Wednesday, April 29, 2020

How much diesel to run a railroad?

Let's play the math in figuring out how much oil it would take to make a seven-hundred mile run with thirty boxcars and a stop every twenty-five miles.

Modern diesel engines have a peak fuel efficiency of approximately 0.4 pounds of fuel per horsepower hour. For example, a 300 horse-power Cummins diesel engine uses approximately 17 gallons of #2 diesel per hour.

The weight of an empty boxcar is approximately 60,000 pounds. Assume a smoldering economy like in the Quest story would half fill it...that is, another 60,000 in cargo.

It takes 1.25 horsepower-hours to accelerate that 120,000 pounds to 25 miles per hour which is probably a reasonable maximum speed before large amounts of time have been invested in bringing the tracks back up to snuff.

Assuming 30 boxcars, then it takes about 40 horsepower-hours to accelerate the entire string to 25mph.

Waving our hands to make inefficiencies disappear, the locomotive must put out eight minutes of 300 horse-power to bring the train up to speed and will consume 2.3 gallons of diesel while doing so.

Since there are roughly 28 stops along the way, there will be 54 "starts" or about 120 gallons of fuel used to overcome the inertia of the train for a round trip.

This is where we throw a fudge-factor at the inefficiencies we blithely ignored. Double the 120 gallons to an even 250 gallons of diesel for a round trip.

Mid-Western oil production
Looking at just the oil production in Indiana:

Indiana was producing approximately 120,000 barrels of crude oil per MONTH over the last few years. That is 60 million gallons a year.

Let's make some crude assumptions. Let's assume they can get 5% of production back on-line and let's assume they can separate 20% into products by simple fraction distillation.  That pencils out to 600,000 gallons of "diesel" fuel a year. Running our tiny railroad two round-trips a month would use 6,000 gallons of fuel or 1% of our calculated production capacity.

Freight versus personal transportation
The states of Illinois, Ohio, Indiana and Iowa have a combined population of approximately 35 million. In a typical year, they use about 5 billion gallons of gasoline for personal transportation.

Our 6,000 gallons per year divided by the current usage of 5 BILLION is one part in 830,000.

So the assumption "There will be no oil" is true for current consumption habits involving use of 4000 pound vehicles driven 10,000 per year in a post-Ebola economy.

That same assumption is probably false if the consumption habits shifted to very minimal use for freight carried by steel wheels on steel rails AND some of the currently producing wells can be restarted. Local production would not be stressed to produce 6,000 gallons of diesel a year. Remember, we only looked at Indiana production and we only assumed 5% would be reactivated.

The thoughts of those in the petroleum industry will be much appreciated. Aggie...that might be you.


  1. A 300 hp engine in a dodge truck gets about 15 miles per gallon. driving 75 mph for 1 hour puts you 75 miles down the road. At 15 miles per gallon fuel consumption you would use 5 gal not 17 gal.

    1. It is not putting out 300 hp to push your truck down the road.

  2. The Electromotive diesels (V-16) that were developed for the modern locomotives are also very typically found powering offshore drilling rigs. They're nice to have becuase they are fully modular and easy to change out parts on one cylinder at a time. You have to use EMD theory on trains because there isn't a clutch in existence that would allow that much slippage on starting - hence electric motors are used in direct drive, with the EMD's being the prime mover to the generators. Early EMDs were about 1,000 hp, modern turbocharged ones are now ~1500 hp. I think EMD locomotives have 2 EMD gensets, but I'm not sure.

    On the refining side, that's going to be a much harder problem to solve. Refineries are tailor-made for their feed stock crude, and that usually is a blend of many different crudes, selected for their natural fractions. It's going to quickly become very complicated to get a refinery up and running, because they are also designed for specific rates of throughput. It simply isn't feasible to run one at some small fraction of its design rate. And the supply stream from the wells - how is that getting to the refinery? Pipeline? Train? Truck? Who is working over the wells to keep them flowing? Who is producing / pumping? What kind of shape are the pipelines in? Might be a theme for you, there - remember Saddam's exit in Kuwait?

    Trains are hugely efficient cargo movers compared to trucks (pound-for-pound), mostly because the frictional (steel on steel) and gravity (grade) losses are well managed. EMDs are thirsty creatures though. I'm no expert but IIRC I would figure on 150 gal per hour per motor at least if they're working hard. It easily takes that to keep an offshore rig running. Just to keep the lights on, these things idle along at around 15 gph - again, IIRC.

    1. At 0.4 lbs fuel per horsepower-hour, that is about 17.5 hp-hr per gallon. 1500 hp for an hour would run in the neighborhood of 86 gallons.

      1500hp could get 30, partially loaded cars up to 25mph in a bit less than two minutes. That is a little bit of over-kill.

      The net result is the same amount of fuel to get to terminal speed but more waste as it idles during coast-down. My assumption is that the train would run up-to-speed and coast down to 80%, then power back up.

      I think there are smaller, shuttle engines for switching yards that are in the 300hp range.

      Regarding refineries, petroleum is a cut-throat commodity business with tight margins. If the goal is to produce some diesel-like fuel rather than the optimum mix of profitable products then more primitive techniques and smaller towers can be used.

      Just my two cents.

  3. Your average tracked excavator uses 12-15 gallons per hour to run at a working RPM. I would say that your 150 gallons per hour per motor is very realistic.

  4. The real issue will be restarting the refinery. That assumes critical personnel did not die, and that the system was purged before shutdown.

  5. I found another data-point. CSX claims to move a ton of freight 492 miles for each gallon of fuel.

    Using the same numbers from above (30 cars with 60,000 pounds of cargo and 1400 mile round trip), the CSX number suggests a fuel consumption of 2600 gallons. That is about ten times my calculations.

    Data wins. There would be some increases in efficiency with the slower speeds but some losses in efficiency with partially filled boxcars.

  6. Have you read the 1632 series of books? They face some of these same questions (though they don't go as far into detail). The Destroyermen series also does it and goes into more detail.
    If possible, converting to natural gas would simplify refining and purification problems. If the gas is 'wet', the liquid could be used with little processing.

  7. Mention of trucks sparked a thought that if the first trains will have only a handful of boxcars, then perhaps off the shelf trucks will be used for the motive power after they are fitted with a wheel conversion setup.

    The ones I've seen keep the rubber tires, and they have smaller steel wheels.

  8. Build wagon beds for handcars and pull them with horses.

  9. I don't think you can run a refinery for very long at 10% or less of capacity.

    Nor, I think, could you keep it fed for long in your world, as that would require lots of people to keep the oil flowing (even at a reduced rate) in the pipeline from source to refinery. Further, pipelines don't work well at reduced capacity either.

    Your HP and fuel use numbers are pretty much correct. I'm not sure you would even need that much HP due to the "gearing" provided by the genset.

    Also, remember that these diesel locomotives are generally hard to start, so they effectively never shut down. I am not sure what their idle fuel consumption is. Perhaps some work it could do while sitting? Compress air into tanks or pump water electrically while idling? Something to make it more efficient?