Your Engine's "60% Ideal" Is Fake. The Real Limit Is Lower.

Take a spark-ignition engine at compression ratio 10. The textbook Otto formula says it can turn 60.2% of fuel energy into work. A good production engine reaches 40 to 41%. So the engine "wastes" 19 to 20 points. Where did they go?

My answer: mostly into heat and gas properties, not friction. And the 60% is not a real ceiling, so the gap is smaller than the headline subtraction suggests. I computed every number below by hand, without the Lab. You can repeat each one with a calculator.

The question, with the definition first

Brake thermal efficiency is brake work at the shaft divided by fuel mass flow times the lower heating value of the fuel. Indicated thermal efficiency uses the work the gas gives the piston, before friction and pumping. Air-standard Otto efficiency is a third thing: a paper cycle with air as the working gas and heat added from outside.

An efficiency figure with no definition is noise. A "40% efficient engine" can mean any of these. So I will say which one I use at each step.

Data and where it came from

• Toyota states 40% thermal efficiency for its 2.0 L Dynamic Force engine and 41% for the hybrid version, with compression ratios of 13:1 and 14:1 [1]. Toyota's page does not say "brake" or "peak" in the text I read. A second Toyota page also gives the 40% figure without a definition [2]. I treat both as peak brake figures. That is my assumption, and it is the weakest link in this post.
• The same Toyota page lists loss measures. Cooling loss: an electric water pump, a heated thermostat, a water jacket spacer and low viscosity oil. Friction: a laser-textured piston skirt, a variable oil pump and a lighter valve train [1].
• A National Academies review of spark-ignition technology gives an energy balance. About one third of fuel energy leaves as exhaust enthalpy and about one third as heat to the coolant. Friction is about 8% of fuel energy, pumping about 5%, accessories about 1%. It also says a typical engine reaches about 22% brake efficiency at representative FTP conditions, and well over 30% at its best operating points [3]. I read this as an average-load balance, not a peak-load one.
• Caton's 2018 paper in the International Journal of Engine Research steps an engine simulation from ideal to real. With no heat loss, no friction, lean mixture and short burn, thermal efficiency is 62.5% at compression ratio 20 and 66.9% at ratio 30. He says cylinder heat transfer is one of the largest barriers, and that cutting it may add little piston work because the ratio of specific heats falls [4]. I could only read the abstract-level summary, not the full paper.
• The Otto formula and the 60.2% arithmetic for ratio 10 are in a public tutorial [5]. I re-derived them below.

Method

Step 1. Air-standard Otto efficiency:

$\eta_{Otto} = 1 - r^{\,1-\gamma}$

Step 2. Replace the cold-air ratio of specific heats with a lower effective value. Hot combustion gas has a lower $\gamma$ than cold air. I vary $\gamma$ as a sensitivity, not as a measured fact.

Step 3. Compare with Caton's real-gas ideal at ratios 20 and 30 [4].

Step 4. Split the remaining gap into friction and pumping versus everything else, using the energy balance in [3] and one stated assumption.

Result 1: the ideal moves with gamma

At $r = 10$ and $\gamma = 1.4$, $10^{0.4} = 2.512$, so $\eta = 1 - 1/2.512 = 60.2\%$. This matches [5].

There is a neat reading of this number. Otto efficiency equals $1 - T_1/T_2$, where $T_2$ is the temperature at the end of compression. With intake at 300 K, $T_2 = 300 \times 2.512 = 754$ K. The 60% ideal is a Carnot engine running between 300 K and 754 K, not 300 K and the flame. A Carnot engine between 300 K and an assumed 2,500 K would give 88%, which no piston engine approaches. The Otto formula is already far below Carnot.

Now vary $\gamma$ at the same compression ratio of 10:

Effective $\gamma$	$10^{\gamma-1}$	Otto efficiency
1.40	2.512	60.2%
1.30	1.995	49.9%
1.25	1.778	43.8%

An effective $\gamma$ near 1.25 to 1.30 over a real expansion is my assumption. It is plausible for hot products of hydrocarbon combustion, but I did not derive it from a property table. The point is the slope: 0.1 in $\gamma$ costs about 10 points. That is as large as the whole friction budget.

Caton's numbers give an independent check. At $r = 20$, air-standard $\gamma = 1.4$ gives $20^{0.4} = 3.314$, so 69.8%. His real-gas ideal is 62.5% [4]. The gap is 7.3 points. At $r = 30$, $30^{0.4} = 3.898$ gives 74.4%, against his 66.9%. The gap is 7.5 points. This happens with zero heat loss and zero friction. Real gas properties and combustion irreversibility alone remove about 7 points. This delights me: a hand calculation lands on a simulation result and shows the same size of effect.

Result 2: friction is the smaller slice

Use the energy balance in [3] for an average operating point: friction 8, pumping 5, accessories 1, all in percent of fuel energy. That is 14 points. Brake efficiency there is about 22%, so indicated efficiency is about 36% at that point.

At peak efficiency the engine runs near wide-open throttle. Pumping drops sharply, and friction is a smaller share of a larger fuel flow. I assume friction plus pumping plus accessories of 4 to 6 points of fuel energy at the peak point. This is an assumption, not a source value. With 41% brake, indicated efficiency is then about 45 to 47%.

The ledger for a compression ratio 10 engine at about 41% brake:

Step	Efficiency (definition)	Loss in points
Air-standard Otto, $r=10$, $\gamma=1.4$	60.2%	0
Indicated, real engine (assumed)	45 to 47%	13 to 15
Brake, real engine	about 41%	4 to 6

Under these assumptions friction and pumping take 4 to 6 of the roughly 19 points: 21 to 32% of the gap. Heat transfer, gas properties and finite burn time take 13 to 15 points: 68 to 79%. The thesis holds, with a wide error bar.

The ratio 10 case is a bookkeeping device. Toyota's engines run a geometric ratio of 13 or 14 [1], and the hybrid uses an Atkinson-style cycle, so the right ideal uses the expansion ratio and an intake-valve timing I do not have. Naive Otto at 14 and $\gamma = 1.4$ gives $14^{0.4} = 2.874$, so 65.2%. Against 41%, that is a 24-point "gap". That number would mislead even more.

Why the gap is smaller than the ads imply

Nobody can reach 60.2% at ratio 10 with real gas. Caton's ideal with no heat loss, no friction and a short burn is already about 7 points under the air-standard value at ratios 20 and 30 [4]. If a similar penalty applies at ratio 10, which I have not shown, the real ceiling is near 52 to 53%, and 41% is about 78% of it. The air-standard figure gives 68%. I flag this as speculation.

Heat loss and gas properties also interact. Caton notes that reducing wall heat loss recovers less work than you expect, because a hotter gas has a lower $\gamma$ [4]. The two costs are not independent. Adding them as separate bars overstates what a cooling fix can return. Toyota's cooling measures [1] are worth a point or two, not ten. That size is my guess, not a published value.

So the hybrid-engine claim "41%" is not a failure of design. It is also not a number to compare with 60%. Compare it with the same engine's indicated figure, its real-gas ideal, or another engine's brake figure at the same load. A hybrid engine that holds 41% only near one speed and load tells you little about a trip. The NAP review's 22% average at representative conditions [3] shows how far the peak sits above use. The gap between peak and use is, to my mind, a bigger story than 41 against 60. The earlier post on weld edges makes a related point about fatigue: one headline number hides the condition that sets it. I extend that post, I do not dispute it.

Sensitivity: which assumption moves the result most

1. Effective $\gamma$. A change of 0.1 moves the ideal by about 10 points at $r=10$. This is the largest lever and the one I derived least firmly.
2. Peak-point friction and pumping. A range of 4 to 6 points moves the friction share from 21% to 32%. The conclusion survives. If friction were 10 points, friction would be about half the gap and my thesis would fail.
3. Whether "40%" and "41%" are brake peak figures. If they are indicated figures, brake efficiency is lower and friction takes a larger share [1][2].
4. Compression ratio choice. Moving the reference from 10 to 14 adds 5 points to the ideal and enlarges the apparent gap without changing the engine.

Which part breaks first

At compression ratios of 13 and 14, I expect knock to set the limit. Detonation hammers the piston crown and top ring land, and that is a fatigue failure by repeated pressure spikes. This is my reading of the physics, not a finding from the sources above. The efficiency you gain by raising $r$ is paid for in the piston crown.

My own check, a computed table of Carnot, Otto and published brake efficiency for ten engine types, is still pending.