In other words, all other things being equal, what would the miles-per-gallon be for a car with an internal combustion engine that was running at 100% efficiency?
To little information to say.
The MPG of a car is determined by the weight, rolling resistance, air resistance and losses in the drivetrain, among other things.
Currently the best experimental cars have achieved over 300 MPG. This car achieved something like 12,000 MPG equivalent on Hydrogen.
Without more information on what counts as a “car” and what sort of driving cycles you use to compute MPG, this question is a bit like “how long is a piece of string?”
Still, you might be interested in the Automotive X-Prize, which was a recent competition to maximize efficiency using plausibly useful sorts of cars. One of the winners, the Edison2, achieved 110 mpg (this is some sort of “mixed city/highway driving” figure). It’s a very lightweight, streamlined car that runs on a small-but–ordinary internal combustion engine.
(Also complicating your question is definitions of “efficiency”. In terms of thermodynamic efficiency, heat engines are stuck below hard theoretical limits. Real-world power plants achieve something like 40% efficiency, cars closer to 20%, and the theoretical best thermodynamic efficiency isn’t a lot higher.)
Yeah, I know it’s a muddy question but I’m not sure how to word it exactly.
Fair enough – best way to fight ignorance is to ask questions and learn things along the way!
This is a good page that describes the energy losses in a typical car, and (optimistically) where there’s room for improvement.
The biggest energy loss (62% engine loss on that page), however, is largely due to unpleasant laws of physics. Better engine design can nibble that figure downwards, but no internal combustion engine will be very efficient.
Als, that cite is a little confusing in the way they use percentages to mean different things. A diesel may be 35% more efficient than a gasoline engine, but that means a diesel’s thermodynamic efficiency might be 40% compared to the gasoline’s 30% (and I’m using ballpark sorts of numbers here).
It is worthwhile to expand upon this for those unversed in thermodynamics. The work you get from an internal combustion engine comes from the energy released during the combustion process. In other words, you inject fuel into a chamber, add in some oxidizer (the oxygen in air, or for high performance race cars, bursts of nitrous oxide), compress the mixture to a near optimal pressure, and then ignite it, either by electrical impulse (spark plug) or overcompression (dieselling). The resulting combustion produces a very hot expanding gas that pushes against the reciprocating piston, which transfers the omnidirectional momentum of the gas to a directional impulse. This energy loss causes the gas to cool, and then the next compressive stroke pushes the resulting products into the exhaust system (for a four cycle engine). This is all described in theory as the thermodynamic cycle of a heat engine, for which the ideal is the Carnot cycle. As can be seen from the diagram, the amount of work that can be developed, aside from any frictional or thermal losses, is related to the difference between the hot temperature (just after combustion) and the cold temperature (at exhaust). The hot temperature is governed by the combustion temperature of the products, and the cold temperatre is, at lowest, the ambient temperature. This gives the amount of work that it is possible to develop for a given entropy change in the engine, as seen in this T-S diagram. The white is the amount of work developed, and the pink is the energy developed in the system that cannot be extracted at the low temperature, T[sub]c[/sub].
Why can’t we get perfect use of the thermal expansion of the gas? Becuse in order to do so we would need to let the gas expand until it cools to the ambient temperature. This isn’t practical for two reasons; for one, it would take a very long (effectively infinite) piston stroke with negligible friction or other resistance to extract that energy, and two, we would need a completely adiabatic (perfectly insulated) chamber. Both of these are neither physically possible, nor practical to implement in a compact engine Instead, we are satisfied to extract the amount of work that can be developed over the short period of a piston stroke (~0.05 to 0.1 second interval). This also limits the amount of heat energy lost through the engine block, such that thermal losses are essentially negligable in terms of engine output (although the transient losses at the cylinder walls can and do affect the completeness of combustion).
What can be done with internal combustion engines to approve efficiency? Aside from reducing friction losses (microfinishing bearing surfaces, improving transmission efficiencies) and ensuring nearly complete combustion (multiple intake/exhaust values, variable cam positioning to achieve ideal compression at different engine speeds, obtaining higher compression through the use of an external pressure booster) we could increase the thermal gradient, which increases the amount of work that can be extracted for a given entropy level (see the second link above). This can either be done by raising T[sub]h[/sub] or lowering T[sub]c[/sub]. In practice, lowering the rejection temperature isn’t realistic for the reasons stated above. Raising the temperature would require different propellants and an engine structure that could withstand the higher temperatures, pressures, and dynamic environments associated with a higher performing fuel and oxidizer. We do this with rocket systems, using cryogenic oxidizer (liquid oxygen) and highly sometimes reactive fuels such as liquid hydrogen, but only by pushing materials science to the limit, and then, only on engines that are extremely costly to build and operation for a tiny fraction of the lifetime of an automotive engine.
The modern automotive engine is near the limit of what can be conceivably extracted from petroleum fuels (gasoline or kerosene), which aside from being (currently) readily available, are also stable at normal temperatures, have high energy density, and are only moderately flammable in comparison to higher performing fuels. Other practical fuels like methane, ethanol, or methanol, have lower energy density and also don’t function as well across the same temperature spectrum. Until someone develops a practical and cost-effective higher density source of energy storage (which neither hydrogen fuel cells nor electrochemical cell batteries are even within an order of magnitude of petrofuels) we’re stuck with the <40% thermodynamic efficiency of an internal combustion engine.
Stranger
Also - many cars are overpowered and there is a loss of efficiency from driving a too-big engine. However, the smaller the engine, the poorer the acceleration. If you are willing to put up with the smallest engine and painfully slow acceleration (and slow cruising, to reduce wind resistance) then you will get better efficiency.
Another way to improve efficiency is to reduce rolling resistance by overinflating the tires. After all, they are only at that pressure for comfort and tire life.
the other problem is that just referring to "peak"efficiency is not terribly useful. a reciprocating piston engine’s peak efficiency is only attainable when it’s running at the RPM where it makes peak horsepower and is at wide-open-throttle (WOT.)* Nobody drives like that, so in most cases of everyday driving the efficiency of a gasoline engine is rather lower than peak due to pumping losses caused by the nearly closed throttle plate.
- obviously this doesn’t apply to diesels due to their traditional lack of a throttle.
I see. Thank you!
it’s not really possible to ask the question in a meaningful way. there are far too many variables in play.
As speed approaches zero, efficiency in MPG or equivalent units approaches infinity. To get a meaningful answer, you need to specify a minimum speed.
And I don’t think it’s all that useful to consider energy efficiency (as opposed to mileage efficiency), since all cars ultimately have the same energy efficiency: 0%. If you drive somewhere and then return home and park in your driveway again, the net work done is zero, but you’ve burned some gas. But the measure of usefulness of a car isn’t work done, it’s total distance traveled.
Um…no; this is extraordinarily bad advice. Tires that are overpressurized for their application will suffer from a loss of elastic response, especially in sidewall dynamics that will adversely affect roadholding capability and recovery. It is, by analogy, the different between walking around in rubber-soled hiking shoes and stiletto heels. This renders the vehicle unsafe from a controllability standpoint, not to mention more prone to blowout and bead separation failures.
In practice, the energy loss due to tire flexure is very small, and for ordinary passenger car tires, almost entirely acoustic in output, which is not affected by tire pressure but instead by tread configuation (e.g. big knobby tread versus streamlined tread). These losses are entirely by design; that is, the tire serves as the first level in your suspension system and uses the natural hysteresis in the elastomer material and geometry of the tire to provide a controlled degree of compliance to the surface, ensuring positive traction. If you had tires with no compliance, the vehicle would skid around like a newborn colt even on dry pavement.
The modern automotive tire is one of the most highly engineered single mechanism devices in existence and is optimized to function at the specifed pressure range. You should never inflate your tires higher than the maximum recommended tire inflation pressure, and there is really very little reason to inflate them higher than the manufacturer’s recommended pressure unless you are driving a different tire configuration than the OEM tires e.g. low sidewall tires on a car that is spec’d for normal height sidewalls.
Stranger
I can well believe that. It is a facinating concept. Worthy of it’s own thread. 
What other devices fall into the category of highly engineered single mechanism device?
Tires as noted. Bearings? Drugs probably don’t meet the definition of single mechanism.
I will start the thread unles Stranger would be so kind, but not being an engineer I don’t have much to contribute.
Still, a facinating concept.
Thanks for the thorough explanation. I remember just enough thermo to get myself in trouble if I’m not careful. I can take a stab at explaining general principles, but I leave more thorough explanations to more knowledgeable dopers like you.
True, though engineers design the engines (and transmissions) so that they’re operating near peak efficiency at cruising speeds.
Incidentally, hybrid cars can ALWAYS run a small engine at peak efficiency, or just cut it off. That’s one of the tricks that makes them more efficient.
Yeah, there’s no direct way to translate between energy efficiency and MPG, but I thought it was useful to bring up in the accounting of all energy losses. The biggest energy loss is heat, and any improvements here will have a significant impact on MPG. There’s just not a whole lot of room for improvement remaining.
So in the end there are a few basic approaches to increasing MPG. Increasing the engine efficiency helps, but it’s already damn near optimal. Decreasing vehicle weight pretty increases MPG by reducing the amount of energy needed to accelerate the car and also reducing rolling resistance. But given that people don’t buy small cars very often, this means using lightweight materials like carbon fiber, which currently is the stuff of supercars. Streamlining helps reduce drag, particularly improving the efficiency at highway speeds. Take those approaches as far as possible and you get a car like the previously mentioned Edison2.
Hybrids and electrics have one additional trick, regenerative breaking. That allows the recovery of a small but useful amount of energy. But that leads to another tradeoff, since batteries are very heavy.
Peak economy, not efficiency. even at 70 mph I guarantee you your engine is not running at peak efficiency with the throttle barely cracked open.
I thought that a speed of between 45 to 55 mph was the “sweet spot” where you achieved maximum efficiency. I can’t imagine that cruising at 1mph would give you better efficiency. You are losing heat energy with the engine idle.
The cite of peak gas efficiency (which is at about ~45 for the cars of that era) was based on a study performed during the Carter Administration, and was dominated cars using very inefficient automatic transmissions, not optimized to reduce air drag at highway speeds, using fixed timing and carboration instead of variable valve timing and fuel injection. I don’t know what the average peak MPG is today, but I would guess that it is in excess of 60 MPH.
However, your nitpick is essentially correct. At ~0 MPH, you are consuming fuel at idle and making essentially no distance, so your MPG is dismal even though fuel consumption is small. At low speeds you will lose proportionally more energy to rolling resistance and other quasi-fixed losses. At some speed (which depends on the particulars of the car) you will achieve a threshold after which your gains exceed the losses. The same is true of modern jetliners, which are more efficient flying at higher altitudes at high subsonic or transonic speeds (>0.8 Mach) than they are at lower speeds.
Stranger
Thanks for the detailed explanation. Given all this, it’s impressive the efficiency gains auto-makers have managed in recent years. I bought my first car in 2000, and it was eight years old at the time. It had a 1.25L engine, put out 75 BHP, weighed 860KG and got 41 MPG (official combined figure, consisting of a mix of urban and extra-urban driving). My 2008 model car has a 1.6L engine, puts out 115 BHP, and manages 44 MPG despite weighing much more at 1,250KG. In the real world, I get about 42 MPG out of it with my driving patterns, while I got high 30s in the old one.
I’m pretty sure this is incorrect. Looking at my trip computer, I get much better mileage at 40 MPH than at 70. This isn’t surprising, as air resistance increases as cube of speed.
One thing which can be definitively said is that each car and driver are going to have very different “optimal” speeds for cruising. And sometimes the results seem rather strange, as we found out in this column.
One small nitpick: Only the very simple “diesel” engines used on model airplanes work as described above, with fuel and air mixing prior to compression. These are known as semi-diesels, and rely on ether in the fuel to permit compression ignition at modest compression ratios. True diesels do not fill the combustion chamber with a fuel/air mixture.
The key thing is that a true diesel injects it’s fuel only after the air has been compressed to the degree needed for spontaneous combustion. Ideally the fuel would burn instantly, and at the rate it is injected. This allows control of the ignition timing, and is also what prevents very low octane diesel fuel from detonating at compression ratios that are about twice those of spark ignition engines. It also means that the fuel/air ratio is largely irrelevant. With most diesels, you can keep adding fuel and getting more power right up until the exhaust valves fail from oxidization at high temperature .(burning) even with the cooler exhaust of a diesel, the valves won’t take the heat that they will in a gas engine, because the exhaust still contains plenty of unburned oxygen.
In practice there is a short delay between the start of the injection, and ignition. This allows the first part of the injected fuel to detonate, and this is what gives a diesel engine it’s characteristic knocking sound. Contrary to spark ignition engines, the more quickly the diesel fuel self ignites from compression, (higher cetane rating) the less noise the engine will make, because less fuel will have been injected before ignition begins.
To relate this to the OP, this injection controlled burn rate allows diesels to operate at very high compression ratios that permit high ignition temperatures and allow a lot more energy to be extracted from the hot gas prior to exhaust. The materials of the engine are what limit the compression ratio, both in terms of strength, and ability to withstand heat. The strength issues can be attacked somewhat in stationary diesels just by making things beefier, but this adds weight that you don’t want in a vehicle. If there were a breakthrough in strong, high temperature, low friction materials, you could see a great improvement in Diesel efficiency. Certain ceramics seem promising, but have been for a couple of decades. Believe it when you can buy one off the car lot.