Spcifically why Are Tier 4 Locomotives Less Fuel Efficient

10 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    September 2013
  • 2,299 posts
Spcifically why Are Tier 4 Locomotives Less Fuel Efficient
Posted by caldreamer on Friday, September 17, 2021 1:38 PM

Please explain specifically why the tier 4 locomotives are less fuel efficient.


  • Member since
    December 2017
  • From: I've been everywhere, man
  • 4,021 posts
Posted by SD70Dude on Friday, September 17, 2021 2:57 PM

For the same reasons that the smog control equipment found on 1970s era American cars resulted in lower fuel economy.  Even catalytic converters reduce engine efficiency, though far less than some older equipment and their benefits far outweigh the loss.

The current GE and EMD Tier-IV locomotive engines use a technology called exhaust gas recirculation (EGR), which is exactly what it sounds like.  Some exhaust is sent back into the air intake where it of course displaces clean air.  This results in less oxygen being available to burn the fuel and a lower temperature within the cylinder.

I'm not sure if the locomotive engines also use diesel particulate filters (DPFs), or how the cleaning/regeneration process works if they do. 

The reason behind this is the need to reduce the production of nitrogen oxides (NOx) in order to comply with emission regulations.  The higher the cylinder temperature, the more NOx is formed.  NOx is but one of several pollutant categories that the emission standards focus on, and tuning the engine to fall within the limits of each category is a fine balancing act, as the settings that reduce some pollutants increase the production of others. 

Having less oxygen available in each intake charge means the fuel does not burn as completely, resulting in lower overall engine efficiency and somewhat increased production of soot, unburned hydrocarbons and carbon monoxide.

Greetings from Alberta

-an Articulate Malcontent

  • Member since
    January 2019
  • 1,224 posts
Posted by Erik_Mag on Friday, September 17, 2021 10:09 PM

In general, the higher the peak temperature, the more efficient an internal combustion engine will be. As SD70Dude pointed out, high peak temperatures also generate more NOx - I recall my engineering thermo prof stating that 2700F was where NOx formation really takes off. The prof had also mentioned that retarding the spark on a gas engine also reducs NOx, but at the cost of power and efficiency.

  • Member since
    September 2003
  • 19,026 posts
Posted by Overmod on Saturday, September 18, 2021 7:59 AM

The issue with high peak temperature is partly associated with engine speed (the faster it is, the shorter the interval for completion of combustion) -- this is less of a concern for a 900-1050 range.  EGR is also somewhat less of an issue on diesels than on spark engines that don't run under polynucleate-ignition conditions, but there is still the time involved at ignition temperature for the diluted oxygen to react with the fuel charge.

I think some of the engines do have DPF and associated fuel burn for regeneration.  It is just as idiotic and feel-good a method of 'pollution' control as on smaller road vehicles.  It would certainly account for lower 'mile per gallon' fuel economy, but of course why let a little something like higher carbon-emission mass get in the way of soot reduction.  (The problem here is that visible soot isn't the major health concern of diesel PM; it's nanoparticulates from early quench, which happily scoot through any practicable DPF and the generation of which is enhanced by typical NOx-reducing approaches on turbo engines...)

When FGR/EGR require gas cooling for their use, you start to get the issues Shadow's owner repeatedly mentions.  If the exhaust were nice clean 'spent' gas, there would be comparatively little issue with it.  But often enough to matter it is not, and I have yet to see any system that only enables the EGR when gas quality is suitable.

I also have not seen EGR systems, even on large turbo engines, that do not have some degree of gas cooling (usually separate from formal intercooling arrangements).  All the heat so removed is thermal energy 'bought and paid for with fuel consumption' that cannot be used to improve subsequent compression-ignition temperature rise...

The thing I continue to find a little astounding is that modern engines that need to use DEF/urea also continue to use wack DFP and regeneration and employ complicated EGR for "NOx reduction" when only slight formal changes in the SCR system would let them address nitrogen oxides entirely post-combustion.  If that is done, compression ratio can be increased to most efficient level permitted by the engine construction (e.g. casting wall thickness or main bearing tribology) and the engine become much more efficient, even as the higher firing temperature and longer duration at high temperature reduce the potential for persistence of nanoparticulate matter to quench.  It would then remain to be seen if there is a cost-effective niche for EGR to be retained for exhaust-heat regeneration rather than misguided pollution control... my guess would be 'no' for practical reasons, but I could also see ways to make it workable.

  • Member since
    April 2018
  • 30 posts
Posted by BDA on Saturday, September 18, 2021 7:07 PM

My take is similar to the above  , mostly .

All internal combustion engines need a heat source and works like this .

Chemical energy/heat energy/pressure energy , push the pistons spin the crank turn the alternator etc .

The whole point is to get the most heat to develop the most torque for the least amount of fuel bought and paid for .

The function of exhaust gas recirculation is to reduce combustion temperatures below the point where Nitrogen Oxide (NO) production really takes off . Just on "NOx" . This is a bone head term used to describe the bonding of 1 nitrogen atom with one oxygen atom . 

Cooling exhaust gas for recirc is not an issue as far as heat or energy loss is concerned . Its about trying to keep inlet charge gasses cool , non cooled EGR probably undoes part of what inlet air intercooling does . The issue with EGR cooling is passing a dirty oily gass through a cooling core and trying not to choke it with coagulated oil soot moisture etc . 

Anyway , anything that reduces combustion temperatures reduces the thermally driven expansion that pushes cylinder pressures up and drives the pistons down their cylinders . Heat loss is power loss , you are basically not getting all of the energy potential for the amount of fuel injected . 

Rail Road operators are not stupid , they pay huge fuel bills so wouldn't be real interested in buying new power with increased complexity , pathetic reliability , AND increased fuel consumption . Why would they bother when they can rebuild older units to say T3 performance/reliability/fuel consumption .

The thing I'd like to know is the costings involved with DEF . It has to be significant otherwise the rebuilding of older power would not be going at the rate that it is . 

  • Member since
    July 2004
  • 2,741 posts
Posted by Paul Milenkovic on Saturday, September 18, 2021 8:59 PM


If that is done, compression ratio can be increased to most efficient level permitted by the engine construction (e.g. casting wall thickness or main bearing tribology) 

The guy I once worked for at a major North American manufacturer of automobiles told me that the limit on compression ratio, whether for a diesel or a "fast burn" gasoline engine that tolerates a higher compression ratio on regular-grade gasoline, is engine friction.  That is, there are diminishing returns in overall efficiency when boosting engine efficiency only to have the increase engine power go into the friction of piston rings and bearings acting against the higher compression pressures.

Don't know, however, if this observation scales to the much larger cylinders in locomotive diesels.

The other thing he told was that the "three-way catalyst combined with stoichiometric combustion (stoichiometric combustion that is regulated by those exhaust O2 sensors monitored by the engine computer adjusting the air-fuel ratio) will restore fuel economy to the pre-emission control level."  This was a prophecy at the time I was told this because the three-way NOx-reducing catalytic converter was still in the R&D phase and only Volvo had this on a production car, but this prophecy was informed by what the R&D people had learned at the major auto companies.  By this, he meant that the three-way catalyst would undo the loss in power and fuel economy from the 1970's era smog controls, and it certainly did once it made its way into production cars.

So exhaust after treatment is indeed the way to go, after treatment of a diesel requires DEF, the trucking industry has embraced DEF as the way forward from those horrid engine-wrecking economy-robbing EGR systems as our colleague commenter here can attest, and why the railroad industry won't go that route is another of life's little mysteries.

As to why DEF is still combined with (some EGR), your current-day car with this three-way catalyst NOx aftertreatment for a gasoline engine still uses EGR, maybe not the high power and economy robbing levels of the 1970s.  There are always engineering tradeoffs and the optimum is sometimes achieved with using both methods.

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?

  • Member since
    July 2004
  • 2,741 posts
Posted by Paul Milenkovic on Saturday, September 18, 2021 9:13 PM

One thing I am curious about is why the railroads even have the option of rebuilding the older Tier 3 units?  I guess in most states, you had the option of keeping your old car, but cars last only so long before it becomes prohibitively expensive -- rebuilding engines or transmissions is usually not an economic proposition on passenger cars.

Do truck operators have the option of rebuilding older trucks that do not meet Tier 4 or whatever rule they have to adhere to?

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?

  • Member since
    January 2019
  • 1,224 posts
Posted by Erik_Mag on Saturday, September 18, 2021 9:41 PM

I've come across a couple of news reports about modifying the injectors in diesel engies to be like Bunsen burners. The first was a report on work being done at Sandia, and the more recent one was a patent applciation or grant from Japan that seemed to describe the same thing. IIRC. Sanda claimed that the device reduced NOx as well a particulates, while not having a negative impact on efficiency.

My guess is that the tubes promote the vaporization and fuel air mixing before initiation of combustion which reduces particulates. NOx reduction may come about when running with excess air, were the excess air cools the flame.


Friction limiting of the compression ratio makes sense, especially when coupled with the extra engine weight needed to handle the higher forces involved. Another limitation on compression ratio is valve lift, which would impact "breathing". Friction can be partly overcome by using longer connecting rods for a given stroke, resulting in less side thrust (Volvo did this for a "low friction engine"). Valve lift issues can be addressed by a more undersquare bore/stroke ratio - the Ford 302 V-8 and the Cummins BT6 both had roughly 4 inch bores, but the stroke on the 302 was 3 inches, while the stroke on the BT6 was 4.8 inches.

  • Member since
    April 2018
  • 30 posts
Posted by BDA on Sunday, September 19, 2021 2:10 AM

Things are very different for a spark ignition petrol burning engine .

NO emissions aside petrol engines are regulated by an air throttle or strangler valve if you like . When you strangle an engine in this way you reduce the amount of air flowing into the cylinders , the important bit here is that the less air in means less to compress so the effective or "dynamic" CR is less than the "measured" by volume - static CR . Lets just say that you cars static CR is 9 to 1 , at part throttle cruise being strangled it may be only 5 to 1 . If the static CR was 11 to 1 the strangled/dynamic CR will be higher . The pay off is more part throttle torque leading to smaller throttle openings and better fuel consumption .

Everything I've read or experienced showed me that fuel octane is the limiting factor with petrol engine static CRs . An engine that detonates is an engine that will cause itself internal damage . Nock sensors are there to sense this and retard ignition timing to prevent detonation and damaged engines . This is also the reason why cars can be run on fuels that don't have sufficient octane to make best mean torque and get best consumption . Retarded ignition timing has a significant effect on combustion efficiency and fuel consumption , but thats better than fried pistons and blown head gaskets .

Oxygen sensors , these were about allowing your engines computer/ECU to run at around stoi or about 14.7 to 1 air fuel ratio . This was supposedly the ideal AF ratio , ie you burn all the fuel and all the oxygen to be in theory neither rich or lean . Cat converters like the ideal AFR because this allows them to get and stay hot to do the job they were intended to . In some cases leaner than stoi is used but combustion temps can't be allowed to get too hot or again engine damage and NO emmissions get out of hand . EGR was one method used to control combustion temps . Another thing with EGR is that adding an incombustable gas to the inlet charge means that cruise throttle openings can be increased so the effort to drag pistons down on inlet strokes against a barely opened throttle valve can be reduced which can aid fuel consumption . 

By rebuilding diesel locos to T3 I meant getting used power built on Tier 0/1/2 technology or whatever you EPA allows in the US and upgrade them to Tier 3 standards . 



  • Member since
    July 2004
  • 2,741 posts
Posted by Paul Milenkovic on Sunday, September 19, 2021 5:25 AM

The reason for the computer-controlled 14.7:1 air/fuel ratio is the three-way catalyst, a newer type of catalytic converter that performed "after treatment" that removed NOx, one of the pollutants that contributes to smog, from the exhaust.  I am not familiar with the details of the chemical reactions, but I have seen the graphs, and the three-way catalyst is only effective in a very narrow range of the 14.7:1 ratio where air and fuel are in their "stoichiometric" ratio of chemical balance.

That Major North American Manufacturer of automobiles was investigating an alternative approach to achieving the strict regulations on NOx emissions without the three-way catalyst.  That gasoline engine used 1) direct-cylinder fuel injection, 2) a piston with a cupped depression where the fuel injection was aimed, 3) a much higher compression ratio but using only regular unleaded fuel, 4) two spark plugs per cylinder, 5) a lean (higher than 14.7:1) air-fuel ratio and 6) a high level of "swirl" on intake and "squish" on the compression stroke, 6) very high rates of EGR, not only to combat NOx but to also reduce "pumping losses" by operating the engine at part load with much less throttle restriction.  

That engine never made it into production, although 40 years later, that manufacturer and others introduced engines with direct-cylinder fuel injection.  Furthermore, some of the "tricks" to reduce the susceptibility to knock must have been incorporated into modern engines, which have crept upwards in compression ratio while operating on regular fuel.

Interestingly enough, that direct-cylinder fuel injection gasoline engine from 40 years ago introduced the problem of emitting fine-particle smoke that Overmod writes about.  Based on that engine, one could say that there is a continuum of solutions to the combustion problem that combine attributes of traditional gasoline and diesel engines.

My informant back then also told me that the "secret sauce" of that complicated engine that never entered production was "the fast-burn engine", which could be applied to a conventionally carbureted or manifold fuel-injected gasoline engine to allow whatever gasoline-engine compression ratio you wanted, but that compression ratio would be capped at 13:1 for the reasons offered above regarding efficiency loss to engine friction.  The man knew a lot about engines, probably a lot from tinkering in the lab with engine dyno experiments that may never have been published in the open literature, but he was a prophet.  Others he worked with stated that he had run experiments at the extremes of "over-square" strokes and "squish" promoting fast burn by combustion chamber turbulence to see where these extremes failed. 

Whereas it uses direct-cylinder fuel injection, the Toyota 4-cylinder engine in the 2021 Camry is a stoichiometric engine using a 13:1 compression ratio on regular gas.  

As to diesels, these engines must run much leaner than gasoline engines so they don't "roll coal" in emitting clouds of black smoke.  The NOx-reducing aftertreatment required a controlled spray of urea (the so called diesel emissions fluid or DEF).  As Overmod points out, the use of DEF allows emission-controlled diesels to achieve pre-control power and emissions, maybe even better power and lower emissions owing to advances in combustion chambers and fuel injection spray patterns and timing.  The story about railroad resisting the use of DEF is that Operations only wants to supply three substances at engine terminals -- #2 diesel fuel, makeup water for the radiator and traction sand.  The claim is that they didn't want the logistics problem of supplying DEF and not running out and getting fined by the EPA.

It is also my understanding that some of the "exotic" passenger locomotives in recent service use DEF, especially those in commuter service with their own specialized engine terminals and service sheds.  

My question about was whether the EPA allows the railroad to rebuild locomotives to less-than-Tier 4, and I guess the answer is yes.  Unintended consequences of rules and government rule-making like the production of sausage in that you dont' want to see everything in the "input stream" and all of that, I guess.

If GM "killed the electric car", what am I doing standing next to an EV-1, a half a block from the WSOR tracks?

  • Member since
    September 2003
  • 19,026 posts
Posted by Overmod on Sunday, September 19, 2021 6:36 AM

I kept trying to post yesterday but Kalmbach kept blanking so I gave up.

The "NOx" is a Government thing, and in my opinion verging on a scam of sorts.  The primary pollutant from combustion is nitrogen oxide, NO.  This is converted in photochemical smog in the atmosphere into nitrogen dioxide, NO2, which is a nasty respiratory irritant among other things and facilitates ozone generation.  The Government's primary concern is abatement of the latter, but as this requires 'attention to the former' to actually reduce, the responses involve "NOx" (the 'x' being a catchall to describe any number of oxygens bonded to nitrogen).  It's snappier and more scientistic than "NO/NO2" and it has that nifty semantic implication of NOxious.

Where the scam aspect comes in is that the reactions in the atmosphere that produce pollutants like PAN require fairly high concentrations of HC, hydrocarbons from improper combustion, which of course in the bad old days of the '60s and early '70s with huge carbureted engines were lavishly present.  The intervening years of practical modulated fuel injection, careful computer monitoring and control even in transient operation, near-pervasive use of catalyst aftertreatment, and increased overall fuel economy and smaller vehicle size have reduced HC emissions enormously... meaning that the aspects other than ozone generation are likewise reduced.  But you'd never guess that from the government emphasis on "NOx" reduction via ever-more-stringent standards-making for engines or other high-temperature combustion sources like cyclone furnaces.

It was my understanding that the reaction to reduce NO necessarily involved the presence of CO at the catalytic site, and that a primary purpose of the 'pre' and 'post'-cat oxygen sensors was to assure that this monoxide would be present to deal with NO, but then not 'make it out of' the converter.  This is not difficult to arrange around 'lambda1' but gets to be fun outside the stoich envelope either on the rich or lean side, especially when the feedback loop involves actions relatively far upstream of the catalyst...

Keep in mind that lean operation in GDI engines can involve almost inconceivably small fuel mass flow.  Some of the original Volkswagen data indicated in the range of 35,000 molecules of fuel per cycleand this at small-engine speeds that drastically shorten the time for 'full combustion'.  I am not surprised that only direct injection gives the necessary control for full combustion without detonation; modern injection can easily command several cycles of pilot injection as well as modulated main injection with very close coordination to cylinder TDC (the alert will also have noted that this relieves much of the concern with resistance due to high peak firing pressure).  (We get into the fun issue of promotion at this point, where a small amount of fuel is injected purely to react to produce more favorable early combustion conditions for subsequently-injected fuel; this is an important reason for the use of up to 5% diesel in many practical natural-gas-burning compression-ignition engines) -- some of the promoter species generated are precisely the sort of hydrocarbons that facilitate conversion of NO to NO2 in the presence of ultraviolet light of the right frequencies, so the unscientific might conclude that high CR firing conditions (both heat and pressure) might also facilitate conversion.  Were there no SCR this might be a somewhat more legitimate concern.

I suppose I should mention at this point that relatively little about historical gasoline-engine practice translates across to locomotive diesels, even the modern 1800rpm engines (typified by the Cat C175 and larger Cummins QSKs, which have been largely sold to various government agencies for passenger service and therefore politically motivated to minimize their NO emissions to 'government' targets regardless of cost or compromised performance).  These are operated at a relatively stable range of controlled rpm (due to the nature of eight-notch MU control), which greatly simplifies the type of FADEC that needs to be applied to them.  In the 'bad old days' GE chose to use extremely long loading intervals to reduce pollution, and even today limits the loading when the engine is 'wiped' from idle to run 8 -- I'd like to think by running the engine up to speed and stabilizing it until the cylinder structure fully warms up before applying electrical traction load to the alternator.  EFI properly 'interlocked' with excitation can achieve this more quickly than a mechanical rack or injection pump with long lines.

What does complicate engine acceleration is that bugbear of Alcos, turbocharger lag.  The EMD system, which runs the compressor as a supercharger in the lower notches, partly addresses the issue, but in going from a positive-displacement (Roots) blower to a centrifugal compressor has made proportional scavenge air a bit more complicated.  Complicating this is that diesels share with Brayton-cycle engines the need to develop high compression pressure before combustion begins, so they need relatively higher 'horsepower' to physically accelerate into a load, so they benefit from steady-state operation and avoidance of anything that begins to 'lug' them slower to where their governor increases fueling to 'catch up'.

One of the showstopping aspects of DEF is the not-always-publicized 'idea' that, to ensure compliance with the laws, an engine that uses DEF is functionally disabled (the clever euphemism I last saw was "derated to zero") if the fluid feed stops for any reason.  This apparently happened to Metrolink at what might have been the most embarrassing time of all, when it died just as it was being moved into position for a grand public ceremony and couldn't be restarted for love nor money.

Railroads are justifiably not interested in anything that takes an engine completely and suddenly off the line if the DEF starts crystallizing in its tank or there is some sort of malfunction in the feed apparatus, particularly in the sort of operating situation that became near-pervasive a few years ago where locomotive consists might have to sit idling or even have units shut down for extended periods of time on the road.  On the other hand, I'm not surprised that there's concern that railroads would exploit any sort of limp-home provision that allowed the engine to run at reduced power without sufficient DEF.

In order for railroads (or OTR truckers, for that matter) to fully accept DEF, it should facilitate more than just politically-related benefits.  To me this clearly implies removing other methods of NOx response that hamper efficient engine performance or result in maintenance expense or consequences, and designing the engine to run best without regard to NO generation and using the 'mandatory' SCR for abatement of all the emitted NO (which, as it is a proportional and relatively fast-loop system to minimize ammonia slip, it is easily capable of being adapted to do.)

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy