According to the internet, an internal combustion engine is powered by the pressure differential when the fuel expands into gas and has a maximum thermal efficiency of 50%. And a steam turbine is powered by the temperature differential when the fuel releases its chemical energy as heat and has a maximum thermal efficiency of 90%.

What would happen if you sunk the heat from the internal combustion engine into the boiler of the steam turbine, instead of just venting it into the atmosphere as a waste product? I'm guessing it wouldn't have a thermal efficiency of 140% but WHY wouldn't it? Does harnessing the gas expansion reduce the amount of waste heat? According to which physical laws would THAT happen?

Why isn't a combined cycle power plant 140% efficient?

  • 1
    $\begingroup$ In big installations, waste heat is harnessed in all kinds of ways. You can use it to bump up the efficiency of declinators for potable water. You can use it for hotel hot water and for space heating and for deicing. The propulsive thermal efficiency of large ships in now around 50% for the largest diesels, and the coolant will run through several additional processes that absorb heat for non-propulsion uses. Both coolant heat and exhaust gas heat is recaptured. $\endgroup$
    – Phil Sweet
    Oct 6, 2022 at 9:27
  • 1
    $\begingroup$ Check out the first and second laws of thermodynamics. Feynman has some good lectures. Also check out Carnot. $\endgroup$
    – Solar Mike
    Oct 6, 2022 at 9:32
  • $\begingroup$ Also check out Formula 1 MGU-H (motor generator unit - heat), it uses some of the heat of the exhaust gases. Why not use it in normal mass-produced cars? Because it turns out to be the "single most complex and costly component of the current cars" $\endgroup$ Oct 6, 2022 at 10:05
  • $\begingroup$ @SolarMike To be clear, I'm aware of thermal dynamics, that is WHY I assumed it can't be 140% efficient. But there has to be a reason that the power outputs of the two engines don't simply add together. For example, all perpetual motion machines will fail because they violate thermal dynamics, but for each of them, there is an actual explanation of HOW they fail. Something like: "Well, the weights on both sides of the wheel will exactly balance each other out regardless of how you arrange them." $\endgroup$ Oct 6, 2022 at 11:55
  • 1
    $\begingroup$ Isn’t it materials and use that has always controlled the output of engines? Any effort at research would show that… $\endgroup$
    – Solar Mike
    Oct 6, 2022 at 15:14

3 Answers 3


This is known as a combined cycle implementation.

This is extremely common on gas turbine electric generator plants. The exhaust heat goes into a heat recovery steam generator, which makes steam to run a secondary generation system.

Other possibilities for the "waste" heat: Greenhouses, drying paper in a paper machine, building heat, etc.

This is considerably more difficult for a traditional piston engine, but the same rules apply. There is limited potential, however, for the intense capital required for this for use in a regular diesel engine, which will only be used in smaller cheaper applications (typically emergency power). Gas turbines have taken over the market, largely because of the advantages of the combined cycle process.

The second process in the combined cycle extracts work from losses of the first process.

So if Process 1 is 40% efficient, we lose 60%. If process 2 is 50% efficient, we can reduce the losses (60%) by 50%, so the total cycle is now 70% efficient.

  • $\begingroup$ Good information. I suppose that answers the tittle question. I guess the real question is "why isn't a combined cycle power plant 140% efficient". I'll update the question to make it more clear what I'm trying to ask. $\endgroup$ Oct 6, 2022 at 14:40
  • $\begingroup$ @LorryLaurencemcLarry, if the steam cycle is then 50% efficient, then you get back 50% of the losses form the first. They multiply, not add. $\endgroup$
    – Tiger Guy
    Oct 6, 2022 at 18:17

It appears you're already aware of the Carnot limit: the efficiency of a heat engine cannot exceed the fractional difference in absolute temperature between its hot and cold reservoirs. In all cases considered here, the cold reservoir is the ambient environment. Therefore, for the steam engine to reach that maximum efficiency of 90%, its boiler must be at at least ten times the absolute temperature of the ambient environment. The only way you could make the exhaust gases from an internal combustion engine that hot is to reduce the efficiency of the internal combustion engine well below 50% (because the enthalpy of the exhaust gases is "waste" from the ICE). Therefore, when the two are operated in series as you suggest, either the internal combustion engine is less than 50% efficient, or the steam engine is less than 90% efficient, or both.

  • 1
    $\begingroup$ In addition, even if the second-stage steam engine were 90% efficient, the useful output it delivered wouldn't be 90% of the first-stage input, but only 90% of the energy that remains in the exhaust gas of the ICE. $\endgroup$ Oct 6, 2022 at 17:16
  • $\begingroup$ So an ICE running at 50% efficiency produces half the heat of what would have been produced if you took the equivalent amount of fuel and burnt it in the boilers furnace? So harnessing the gas expansion really does reduce the amount of waste heat? Pushing the piston cools the gas because the piston is moving in the same direction as the excited molecules? Is that how this works? $\endgroup$ Oct 7, 2022 at 6:42
  • $\begingroup$ @LorryLaurencemcLarry Yes - or, to be more precise, the piston is moving in the same direction as the subset of the excited molecules that strike the face of the piston. To put it another way, taking energy out of an ideal gas (in a non-flow process) reduces its temperature regardless of whether you take that energy out as heat or as work. $\endgroup$ Oct 7, 2022 at 10:58
  1. Your numbers for efficiency aren't consistent. The steam turbine efficiency isn't a cycle efficiency, it's just the efficiency of the expansion machine. The efficiency of the power stroke in a four-stroke ICE is greater than 90 %. You need to treat the steam machine as a cycle and include all the parasitic losses and the thermal losses of other components. You also need to realize the steam condenser would need to be carried on a trailer and would double the aero drag of a car (It's pretty big.)

  2. The efficiency numbers are a design choice. We can build more efficient ICEs if we choose to. But they are pretty cheap. The cost of the iron components in a long block is about the same as the cost of a tire or a starter motor. If you wanted to spend an extra $2 per horsepower on the iron, you could do quite a bit better. But they end up larger and heavier and want a higher grade of fuel. Basically, as engines have gotten better, we just keep feeding them worse and worse fuel. If you want to buy 100 octane gas for your daily driver, we can bump the efficiency quite a bit.

  3. ICEs are really versatile. There isn't any other technology that has the throttle response and wide efficiency band of an ICE. You can beat ICE's efficiency at any single design point, but you just can beat their efficiency over the range of power levels needed over the road.

  4. We can augment these features of the ICE with physically divers compounding, but that's what we do - augment the ICE. We don't use the compounding for shaft power except in a very few instances.

  5. A few gas and diesel engines were compounded or had multiple expansions. They were tried from about 1890 to 1910. Then we just gave up on it as not worth the trouble. Wikipedia has a list of them - https://en.wikipedia.org/wiki/Compound_engine


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.