Octane and gasoline engines


in Economics, Geek stuff, Science, The environment

Smashed wooden wall

I am learning a lot about hydrocarbon fuels from Morgan Downey’s Oil 101. For instance, that the common understanding of the phrase ‘high octane’ is somewhat misleading. In the context of gasoline-powered internal combustion engines, such as those in cars, the octane rating of a fuel refers to how much it can be compressed along with air before it will spontaneously ignite. In these engines, fuel and air are mixed together and compressed in a cylinder. They are then ignited at a precisely controlled time by a spark plug. Cases where the mixture explodes before then are called ‘engine knock’ and are damaging. As such, engines are designed to use fuel above a certain octane number, in order to be confident that knocking will not occur.

When it comes to choosing fuel to buy, this means it is appropriate to use a grade with an octane rating as high as cited in the operating manual of a vehicle. Going higher, however, may be a waste of money for two independent reasons. Firstly, higher octane fuels are more expensive because they cost more for refineries to produce. Unless your engine is tuned to take advantage of the extra opportunity for compression, no additional power will be generated. Secondly, higher octane fuels often contain less energy per litre, because the hydrocarbons that comprise them have less energy in their chemical bonds. As such, a litre of more-expensive high octane fuel likely will not take a vehicle as far as a cheaper litre of adequate-octane fuel.

Octane numbers are assigned based on how a fuel compares to two specific hydrocarbons: isooctane (which is hard to ignite by compression) and n-heptane (which is easy to ignite that way). 90 octane fuel is thus as resistant to pressure-induced ignition as a mixture of 90% isooctane and 10% n-heptane. Some fuels are even better at resisting pressure-induced ignition than isooctane, and can therefore have octane numbers over 100.

In diesel engines, this is reversed. They do not have spark plugs and rely upon the ability of fuel to ignite spontaneously in the presence of pressurized air. In diesel, the cetane number refers to the propensity of fuel to autoignite on compression. Here, a higher number is more desirable.

One other thing I didn’t know about liquid transport fuels is that the fuel used by piston-driven aircraft, such as small propeller planes, still uses tetra ethyl lead to increase its octane rating. This practice has been discontinued in cars both because it interferes with catalytic converters and because it was massively increasing human exposure to lead – a known cause of brain damage. In aviation gasoline, tetra ethyl lead is used instead of alcohols to boost octane. This is because alcohol-blended fuels are less energy dense, more prone to vapour lock, liable to separate at low temperatures, as vulnerable to corrosion. Such aircraft are a relatively tiny share of the total market for hydrocarbon fuels; still, it isn’t particularly comforting to know that they continuously disperse lead on whatever is below them.

Report a typo or inaccuracy

{ 34 comments… read them below or add one }

R.K. November 4, 2009 at 11:42 am

Great photo – very dramatic.

Matt November 4, 2009 at 3:27 pm

Many owners of small aircraft purchase STCs for them – supplemental type certificates – that allow them to legally run on automobile fuel. Not the majority, mind you, but many. The most commonly used avgas is 100LL, the LL stands for low-lead. It is dyed blue in colour.

Something to keep in mind about octane ratings are that there are two different scales, RON (research octane number ) and MON (motor octane number). Finally, there is an average of the two.

In North America, at a gas pump the octane number displayed is the (RON+MON)/2 average. In Europe, it is simply RON. Therefore often people mistakenly think Europeans get higher octane fuel. They don’t (typically), it is just that the scale used is different.

Milan November 4, 2009 at 3:34 pm

Downey’s book goes into far more detail than my post. The 20-page section on motor and aviation gasoline covers standards for sulfur and aromatics; the RON, MON, and PON (the mean of RON and MON) systems for measuring octane; international variations in the application of octane standards; the evolution of anti-knocking agents; the use of oxygenates like MTBE; alcohols and methylated spirits; American gasoline grades; summer and winter fuels; and different blends of aviation gasoline (100LL, 100, 80, 87, 115, and 145).

110LL apparently contains 2 grams of lead per gallon. The less common AvGas 100 has 4 g/gallon.

Milan November 4, 2009 at 3:39 pm

There is also a section on different kinds of jet fuel, such as the JP-5 blend made specially for American aircraft carriers. Due to the dangers of fire and the difficulty of evacuation, JP-5 needs to have a flash point of over 60°C, compared to 38°C flash point of Jet A (used for US domestic commercial jet flights) and JP-8 (used by the US Air Force and also mentioned here).

All told, Downey’s book is impressing upon me just how absurdly complicated the petroleum industry, and what a huge amount of capital and knowledge are invested in it. It seems to me that if renewables began to approach this standard of sophistication, they would be significantly more competitive than they are now.

Milan November 4, 2009 at 3:49 pm

It may also be worth noting that a lot of octane boosters are rather nasty, in addition to being expensive. Lead “can damage nervous connections (especially in young children) and cause blood and brain disorders.” MTBE can contaminate groundwater. Benzene (an aromatic) is a known carcinogen. Etc.

Leaded gasoline is now banned in many places, and a number of US states have forbidden the use of MTBE (though it is common in Europe). Maximum benzene standards exist, often at 1% – 2% by weight.

Apparently, a single plant in the United Kingdom is the world’s sole producer of tetra ethyl lead. That reminds me a bit of how Quebec continues to mine asbestos for export.

Tris November 6, 2009 at 10:32 am

“Going higher, however, may be a waste of money for two independent reasons. Firstly, higher octane fuels are more expensive because they cost more for refineries to produce. Unless your engine is tuned to take advantage of the extra opportunity for compression, no additional power will be generated.”

This is almost true. The reason it is sometimes not true is that modern engines constantly re-tune the ignition timing based on a number of factors, but mostly detonation. The “earlier” the timing is set, in other words, the closer to top dead centre position of the piston the fuel is ignited, the more power an engine can make (closer to top dead centre – higher compression at point of explosion).

You’d think that the fuel grade was a set thing – so long as you have the appropriate grade of fuel in there, you will produce peak power. But not quite – for one there are other things that can contribute to pre-ignition (detonation), like heat (as an engine warms up, that heat gets into the pre-ignited fuel and makes it less resistant to pre-ignition). So, going beyond your vehicles octane rating sometimes means you will produce peak power more of the time. Furthermore, some cars are tuned “hotter” than normal allowable fuel enables them to run- this means they are constantly in a state of having their timing delayed, and power reduced. This seems to be the case with the Subaru Impreza, which the television show “Fifth Gear” tested and found produced significantly more power with a super-premium fuel than regular premium fuel.

Milan November 6, 2009 at 10:41 am

Downey says that engines will tune down compression if they detect knocking, but won’t increase compression when a higher octane fuel makes it possible.

That may not be true of all cars, but it does mean that buying higher octane fuel is a waste of money for those whose cars have this behaviour.

Tris November 6, 2009 at 11:56 am

Well, actually some cars will increase compression when a higher octane fuel is detected, especially if a car is tuned for the European market and also sold in North America, where the premium fuel is not available. A huge number of cars fall in this category.

Or, if you have a car that doesn’t automatically adjust timing to compensate for detonation, you could take your car tuned for regular fuel, advance the timing, and run premium fuel. Accidentally, this is how I was running the Taurus for a summer – we set the timing, though it was alright, and then found that it ran much better on mid grade fuel. Later we retarded the timing so it ran properly on regular fuel – but the power loss was noticeable.

Tris November 6, 2009 at 11:59 am

Also, using the term “compression” the way Downey seems to be is non-standard. The compression of an engine is the ratio between the volume under the cylinder at top dead centre, and the volume under the cylinder at bottom dead centre. This can’t be changed – it’s just part of the mechanical reality of the engine. What Downey means by compression is the different extent the fuel is compressed at ignition – which can be varied by varying the timing (the closer ignition is to TDC the higher the effective compression at detonation). The standard language is not “increasing or decreasing compression” but “advancing and retarding the timing”.

Milan November 6, 2009 at 12:32 pm

To quote Downey verbatim:

Octane ratings are not correlated with energy content. In fact, some high octane rated fuels, such as pure ethanol, have a much lower gallon for gallon energy content compared with regular octane gasoline. High octane rated fuels allow more fuel, more molecules, to be compressed into the cylinder without autoignition occurring. The extra fuel which can be placed in the cylinder can be used to offset any lower energy content of the fuel. High octane gasoline is therefore less efficient (less miles per gallon) but can be more powerful on each engine stroke, which is why race cars use high octane gasoline.

Individual gasoline engines are tuned to compress a specific quantity of fuel before a spark plug ignites the mixture. The tuning is set for a certain octane rated gasoline outlined in the car owner’s manual. Modern gasoline engines will usually continue to run fairly smoothly by using a computer to change the quantity of fuel and spark timing when given a slightly lower or higher octane fuel than recommended. However, in general, a lower octane than the engine is tuned for may risk damage from knocking, and a higher octane than the engine is tuned for may merely be a waste of money as higher octane fuel often contains less energy. As the engine is not configured to compress additional fuel before igniting it, there would not be any power gain. (p.188 hardcover)

One other thing, about European and North American gasoline: In Canada and the US, octane ratings are the average of the research octane number (RON) and the motor octane number (MON). In Europe, Australia, Singapore, and many other places, they simply use RON. “As a rough approximation, one could subtract 5 points from RON to get the equivalent US federal octane rating. (p.189)

Matt November 6, 2009 at 12:33 pm

…especially if a car is tuned for the European market and also sold in North America, where the premium fuel is not available.

Again, octane rating is not the same here as it is there. Our 94 (available at the pump many places in Canada) is their 98. Anything higher than that is a ’boutique’ fuel.

I agree that they see a lot of engines that we don’t, but actually most of those are diesels, or low displacement versions of cars we otherwise do get, ie. 1.4L Ford Focus.

Re compression: it is indeed fixed in almost all engines (exception being experimental ones, google Saab variable compression). A turbocharged car can use knock sensing to lower the boost the turbo delivers, however. This is as close a regular car would get to variable compression.

Milan November 6, 2009 at 12:38 pm

For what it’s worth, hydrocarbon engineer Robert Rapier reviewed Downey’s book and found few factual errors.

There are apparently a couple of issues about the RVP of ethanol and winter gasoline, however.

Matt November 6, 2009 at 1:03 pm

I read that passage from the book, and he doesn’t mention varying compression, which was rightly pointed out to be impossible for most cars.

Tris November 6, 2009 at 1:06 pm

“High octane gasoline is therefore less efficient (less miles per gallon) but can be more powerful on each engine stroke, which is why race cars use high octane gasoline.”

This is really a partial truth. Higher octane means the engine can run higher compression – and higher compression means increased efficiency. Note the catastrophic loss of both power and fuel economy when compression ratios dropped in the early to mid 70s to comply with NOx regulations.

Even though the amount of energy in the fuel is less, since higher compression means higher efficiency, there is no reason to assume miles per gallon will decrease.

Personally, I noticed generally higher fuel economy when running the Taurus on mid grade fuel with advanced timing. I’m not sure if this offset the higher cost of the fuel.

Tris November 6, 2009 at 1:08 pm

Personally, books where people pretend to be experts on how cars work, when they clearly are not this kind of expert (although they certainly might be on other issues), are one of my pet-peeves.

Milan November 6, 2009 at 1:24 pm

Downey doesn’t say that higher octane fuels have lower energy density than higher octane ones. He says that: “[o]ctane ratings are not correlated with energy content” and that high octane fuel “may… be a waste of money.”

It depends on what is in the gasoline, which in turn depends on factors like environmental standards on aromatics and oxygenates, the time of year and RVP standards, and the relative prices of different inputs for gasoline blending.

Tris November 6, 2009 at 4:00 pm

“Downey doesn’t say that higher octane fuels have lower energy density than higher octane ones. ”


Re: “In fact, some high octane rated fuels, such as pure ethanol, have a much lower gallon for gallon energy content compared with regular octane gasoline.”

The lower energy density is the only supporting evidence for the lower MPG of higher performance engines. So, the fact he isn’t making an absolute claim seems to be a red herring here.

Ok, sure, he isn’t making an absolute claim – but that is the claim made in the logical sequence he uses to conclude:

“High octane gasoline is therefore less efficient (less miles per gallon) but can be more powerful on each engine stroke, which is why race cars use high octane gasoline.”

Tris November 6, 2009 at 4:04 pm

In terms of what is a waste of money, the way most cars are built, marketed, purchased, driven, worn out, mis-used, and discarded is just one prolonged series of economically inefficient decisions. In general, even in abstraction of environmental issues, the car industry (especially in North America) disgusts me.

Milan November 6, 2009 at 4:08 pm

If Matt is right in saying that very few cars use variable compression, Downey’s general claim that using an octane above the one listed in the user’s manual is a waste of money seems sound.

Higher octane means the engine can run higher compression – and higher compression means increased efficiency.

The engine can run at higher compression with higher octane fuel but, in practice, does not. If the engine is putting the same amount of pressure into the combustion chamber – regardless of octane – then it is effectively ignoring the lesser risk of autoignition that accompanies using the more expensive gasoline.

Personally, I noticed generally higher fuel economy when running the Taurus on mid grade fuel with advanced timing.

By what mechanism would you explain this, assuming the Taurus doesn’t have a system akin to Saab’s variable compression?

. November 6, 2009 at 4:13 pm

Saab Variable Compression engine

Because cylinder bore diameter, piston stroke length and combustion chamber volume are almost always constant, the compression ratio for a given engine is almost always constant, until engine wear takes its toll.

One exception is the experimental Saab Variable Compression engine (SVC). This engine, designed by Saab Automobile, uses a technique that dynamically alters the volume of the combustion chamber (Vc), which, via the above equation, changes the compression ratio (CR).

To alter Vc, the SVC ‘lowers’ the cylinder head closer to the crankshaft. It does this by replacing the typical one-part engine block with a two-part unit, with the crankshaft in the lower block and the cylinders in the upper portion. The two blocks are hinged together at one side (imagine a book, lying flat on a table, with the front cover held an inch or so above the title page). By pivoting the upper block around the hinge point, the Vc (imagine the air between the front cover of the book and the title page) can be modified. In practice, the SVC adjusts the upper block through a small range of motion, using a hydraulic actuator.

Tris November 6, 2009 at 4:33 pm

Sorry, I should have been more clear about this.

Downey is using the term “compression” to mean how compressed the fuel is at the instant of ignition. The fuel is much more compressed at TDC than when the cylinder is at the bottom of its movement. Varying the moment of ignition varies the compression of the fuel at the moment of ignition – since the compression of the fuel is changing over time as the cylinder goes up and down.

This is why I said that what Downey means by increasing or lowering the compression is, in standard language, expressed by the language of advancing or retarding the timing. This kind of – not strictly mistake, but non-conventional language, is a real sign that he isn’t an expert on this stuff.

Milan November 6, 2009 at 5:02 pm

The fuel is much more compressed at TDC than when the cylinder is at the bottom of its movement.

More precisely speaking, isn’t this the progression:

  • The cylinder is uncompressed but full of air
  • The piston moves, compressing the air
  • The gasoline is sprayed into the cylinder
  • A spark plug ignites the gasoline?

So, the only things that are ever compressed are air and the fuel-air mixture. High octane fuels ensure that combustion won’t occur before the spark, in cases where the air is compressed to a high degree before the fuel gets sprayed in.

Matt November 6, 2009 at 5:02 pm

I’m not trying to be a nuisance, but from the verbatim quote he never actually uses the noun compression (which has a very specific meaning in an engine), but rather the verb compress for which no good synonym exists. Therefore, I think his point is relevant. If I were to pick nits, though, he talks twice of compressing fuel which isn’t exactly correct. You don’t compress the fuel (being a nearly non compressible liquid) but rather the charge air, a mixture of atmospheric air and fuel.

Also, ignition timing is important not just because of the pressure of the cylinder at the time of ignition, but also because of the speed the flame-front travels. If you imagine the flame front as a plane that propagates from the top of the cylinder at the spark plug and moves rapidly downward, you can see that you’ll want it to match the speed of the downward moving piston face fairly closely. As the engine turns faster, the piston will move more quickly, and for the flame front to keep up, the timing has to be brought forward (advanced). All engines will advance the timing automatically as engine speed increases. (Older cars with distributors did this with flyweights that would fly out further at high speeds and adjust when the distributor fired). A higher octane fuel actually has a slower moving flame front, and allows for more advanced base timing than lower octane fuel. Because of this, the piston is exposed to the combustion for a longer period of time, allowing it to extract more energy from the proces.

Milan November 6, 2009 at 5:04 pm

Or is the gasoline mixed with air before the air goes into the cylinder? The animation here suggests that it is.

Matt November 6, 2009 at 5:05 pm

The gasoline is sprayed into the cylinder

Few (although increasingly more) gasoline cars are direct injection. That is to say, the fuel is sprayed into the manifold and is mixed with the air prior to the air entering the cylinder.

A direct injection car will spray fuel directly into the cylinder. Diesels do this by necessity because fuel timing and ignition timing are virtually synonymous (the diesel injection leads to immediate combustion).

Milan November 6, 2009 at 5:12 pm


Thanks for the information and clarification.

The two outstanding questions seem to be:

  1. If high- and low- octane gasoline cost the same amount, would there be any benefit to using a higher octane fuel in an engine than it was designed to use? (This answer may vary by engine type, sensors, etc)
  2. Given that high-octane fuels cost more, could it ever be economical to use a higher octane fuel than an engine is designed for, given the ordinary sorts of octane-boosting chemicals used in different regions? (This may also vary, given that a few places still allow tetra ethyl lead, many allow MTBE, standards for aromatics differ, etc)

I think Downey definitely means to say ‘no’ to the second question, and probably to the first question also.

Milan November 6, 2009 at 5:17 pm

The Wikipedia entry on internal combustion engines makes clear that most of the useful energy that eventually moves the wheels of the car is coming from expanding gasses: “The mixture is burnt, almost invariably a deflagration, although a few systems involve detonation. The hot mixture is expanded, pressing on and moving parts of the engine and performing useful work.”

Matt November 6, 2009 at 5:39 pm

I can only give an anecdotal answer to your first question.

My car’s manual states that I should use 91 RON octane fuel (it actually says “RON” which I find interesting). This is equivalent to 87 (R+M)/2 octane, which is the lowest grade gas offered at stations. I notice that when I use 87, my engine knocks (or pings) noticeably at high loads, which isn’t very good for it. To offset this, I’ve been buying 90(R+M)/2 at Mohawk, which costs the same as 87 at other stations (the octane is elevated by the addition of ethanol). On this fuel, the car is much happier, although I suspect its fuel economy suffers due to the low energy content of ethanol.

So, to me, there is a benefit to buying higher grade fuel despite the fact the manual says 87 is okay. On the other hand, the manual could actually mean that I should be running 91 (R+M)/2 which is also likely because I don’t understand why they’d specify an octane rating on a scale that isn’t used in the country the car was sold in. In this case, I’m running 1 octane point lower than I should be, but the car works fine on it.

Milan November 6, 2009 at 5:45 pm

Perhaps the appropriate benchmark is avoiding knocking, not what the manual says.

To reformulate question 1: “If high- and low- octane gasoline cost the same amount, would there be any benefit to using a higher octane fuel in an engine than is necessary to avoid obvious engine knocking?”

For example, would there be any benefit in you buying fuel with 92 (R+M)/2 or higher?

Matt November 6, 2009 at 5:54 pm

I highly doubt it.

It seems to make a lot of people feel good though. Once I saw a guy stick the red nozzle (94 Octane) into his Toyota Tercel, a decidedly economy car. I just shook my head at the money he was wasting.

Matt November 6, 2009 at 6:05 pm

In addition to the above, it seems that while cars will de-tune themselves when they experience knock from a lower grade fuel, they won’t up-tune themselves if you use say 94 octane in a car that specifies 91.

A friend of mine drives a turbo charged PT cruiser that specifies 91 Octane. Interestingly the manual says that lower grades are acceptable, but that performance will be degraded.

Milan November 6, 2009 at 6:14 pm

If the answer to the first question is ‘no’ it seems highly likely that the answer to the second is also.

I think alcohols are one of the key octane boosters in North America, meaning the fuel is likely to be less energy dense, the higher octane they are.

Tris November 6, 2009 at 7:41 pm

Hey Matt – thanks for clarifying the ignition-timing-flame-front issue, you understand this better than me.

. January 25, 2010 at 8:04 am

“Avgas (aviation fuel) is nothing like the stuff you pump into a car’s tank at a petrol station. The exigences of flight require aircraft engines not only to be more reliable, but also to deliver more power per unit weight than car engines. Any additional weight detracts from the useful payload the plane can carry aloft. Aircraft engines also operate at 70% or more of full power while they are in the air, unlike car engines which spend most of their time on the road cruising at 20% of full power. Aero-engines also have to cope with the drop-off in performance that comes as the plane gains altitude and the air gets thinner.

For these reasons and more, avgas has a much higher octane rating than petrol blended for cars. It is effectively a super-premium grade with an equivalent pump (or “anti-knock”) rating of 100 octane. It also has a much lower, and more uniform, vapour pressure than petrol—to prevent the fuel from vaporising as the air pressure drops with altitude. “Vapour lock” (bubbles) in the fuel-line can starve an engine, causing it to stall and the plane to fall precipitously out of the sky. All told, avgas is an expensive brew costing around $5 a gallon in the United States and $9 or more in Europe.

In blending avgas, distillers have to add tetra-ethyl lead to protect the valve-seats on the exhaust side of the engine from erosion, while raising the fuel’s octane number and preventing the engine from “knocking”. The latter is caused when the fuel detonates prematurely instead of burning in a controlled fashion. Such detonation can destroy the pistons, cylinder linings and valves within seconds.

A fuel’s resistance to detonation is measured by comparing its performance to that of pure iso-octane (an isomer with the chemical name 2,2,4 trimethylpentane). Iso-octane represents the 100-point on the octane scale; a fuel having 90% of the resistance to detonation of pure iso-octane is rated as 90 octane.

Several grades of avgas are available with octane ratings ranging from 82 to 115, but the type most widely used in the United States is 100LL—a low-lead version the 100-octane aviation fuel used previously. Dyed blue to distinguish it from other grades, 100LL contains two grams of tetra-ethyl lead per American gallon, compared with the previous version’s four grams per gallon. Even so, today’s low-lead avgas still contains four times more tetra-ethyl lead than was in the highest-octane petrol (Sunoco-260) ever produced for street use in America.

The problem, of course, is that tetra-ethyl lead is the toxic substance that was banned from use in petrol in 1996 for health reasons. Back then, the Environmental Protection Agency (EPA) granted aviation fuel a reprieve—provided there was continued progress towards a suitable alternative. Over the past decade, the general-aviation industry has examined over 200 different blends in an effort to find a “transparent replacement” for 100LL. In performance terms, nothing has come close, it claims.”

Leave a Comment

You can use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

{ 1 trackback }

Previous post:

Next post: