# Gas steam hybrid engine

Gold Member

## Main Question or Discussion Point

Has anyone heard of a gasoline engine that instead of using a water jacket for cooling, uses a spray of water to cool the cylinder? I was thinking along the lines of installing a bank of water spray injectors around the cylinder to remove the heat, which creates steam which can be further heated by the exhaust gas cylinder and catalytic converter, ultimately running a small steam engine.

I would imagine that the outer surface of the engine cylinder might be finned aluminum as in air cooled engines.

Anyways, I was just curious if a feasibility study has been made along these lines, or if anyone can think of a good reason why I shouldn't build a small research prototype.

ps. I am familiar with the system BMW is working on, and with the Crower's 6-cycle engine.

Related General Engineering News on Phys.org
Ivan Seeking
Staff Emeritus
Gold Member
The amount of cooling required by the cylinder is critical. If you try to raise the temperature of the steam by providing less cooling to the cylinder, you will damage the engine.

From there, the thermodynamic efficiency of such a heat recovery system makes it impractical. Thot-Tcold / Thot in Kelvin ~ 366-310/366 = 15% at most. Factor in mechanical losses and you easily get something less than 10%. So if you get 50 hp worth of heat transfer, less than 5 hp worth of mechanical power can be recovered. You can get this much benefit or more just by keeping the car tuned-up and the tires inflated properly.

You are far better off just buying a diesel engine which is about 30% more efficient [ie. 30% better mileage] Then, run biodiesel.

Last edited:
Gold Member
The amount of cooling required by the cylinder is critical. If you try to raise the temperature of the steam by providing less cooling to the cylinder, you will damage the engine.

From there, the thermodynamic efficiency of such a heat recovery system makes it impractical. Thot-Tcold / Thot in Kelvin ~ 366-310/366 = 15% at most. Factor in mechanical losses and you easily get something less than 10%. So if you get 50 hp worth of heat transfer, less than 5 hp worth of mechanical power can be recovered. You can get this much benefit or more just by keeping the car tuned-up and the tires inflated properly.

You are far better off just buying a diesel engine which is about 30% more efficient [ie. 30% better mileage] Then, run biodiesel.
Well, as always, I can see that my question showed a complete misunderstanding of the complexity of the problem. But you've steered my brain back into reality. Thank you. It's been too many years since I've studied steam tables.

I was hoping that I could bump the temperature of the steam up to around 1170'K, the operating temperature of the catalytic converter. But I discovered here: http://www.efunda.com/Materials/water/steamtable_sat.cfm

that their scale stops at 640'K when the steam pressure reached 3000psi!

That would be quite the engineering feat on my part to build such a device in my backyard to contain such pressure. I'm starting to understand the difficulty of the problem now.

Where did you come up with your number of 366'K? I've now come up with 470'K(200psi) yielding a 34% efficiency.

Perhaps I should start asking questions about a liquid sodium system? boiling point 1155'K.
nahhh... just kidding.

But I do appreciate the 15% number. I was wondering why BMW was stuck there after 10 years of development.

back to the drawing board.

ps. bumping my tire pressure to 10% over maximum rated, and driving like a granny improved my gas milegage by 17%.

Ivan Seeking
Staff Emeritus
Gold Member
Where did you come up with your number of 366'K? I've now come up with 470'K(200psi) yielding a 34% efficiency.
I was using 200 degrees F as the high temp, and 100 degrees F as the low temp. The engine requires a water temp of about 200 F, so in a best case this is the high temp, and the low temp is limited by the typical highest ambient temps. In reality you couldn't run the temp this high or low without having very large heat exchangers, so the actual efficiency would likely be worse than indicated.

I doubt that enough heat energy comes from the exhaust pipe to be significant. The pipe gets hot, but relatively little heat transfer occurs as compared to the forced water cooling system - an open-air pipe is very inefficient for heat transfer. Consider the surface area of the cooling fins on the radiator to the surface area of the exhuast pipe. Then consider where each is mounted. Also consider that the entire pipe does not get that hot.

Last edited:
Gold Member
I was using 200 degrees F as the high temp, and 100 degrees F as the low temp. The engine requires a water temp of about 200 F, so in a best case this is the high temp, and the low temp is limited by the typical highest ambient temps. In reality you couldn't run the temp this high or low without having very large heat exchangers, so the actual efficiency would likely be worse than indicated.

I doubt that enough heat energy comes from the exhaust pipe to be significant. The pipe gets hot, but relatively little heat transfer occurs as compared to the forced water cooling system - an open-air pipe is very inefficient for heat transfer. Consider the surface area of the cooling fins on the radiator to the surface area of the exhuast pipe. Then consider where each is mounted. Also consider that the entire pipe does not get that hot.
I would argue the high temperature, but after 2 hours of searching, I can't find the nominal surface temperature of an air cooled engine.

About the only numbers I've come up with so far is where the heat in an air cooled engine is dissipated: 44% through the exhaust, 12% through the fins, and 8% to the oil. (courtesy of http://en.wikipedia.org/wiki/Air-cooled" [Broken])

If I had the surface temperature, I could design an engine with two water based coolant systems. One operating at a high pressure and temperature, I'm guessing 470'K at 250 psi, and the other operating to generate steam. Very similar, I guess, to a pressurized water nuclear reactor. Might be simpler than my single system of spritzing water directly onto the engine cylinders. Of course the water pump in the primary system would be a bit of a problem. Though I've heard fuel pumps operate fully submersed in their operating fluids. So maybe not.

Last edited by a moderator:
Q_Goest
Homework Helper
Gold Member
Hi OmCheeto,
I'd agree with what Ivan is saying - it's impractical and uneconomical to try and modify a car to recouporate the heat from the engine and exhaust. It could be done but you'd need to do a lot more than put a jacket on your exhaust pipe or squirt water into the engine for all the reasons Ivan has already mentioned.

That said, if you were trying to generate electric power in a stationary application such as for your home or a cabin where utilities weren't viable, you could do things similar to what you're talking about here. I wouldn't suggest spraying an engine with water, but why not simply let water boil against the engine block? Same thing - you're taking saturated steam off the engine. You could do this at slightly elevated pressure and pump subcooled liquid water into the block to replace the boiled water.

The saturated steam could then be passed through a proper heat exchanger heated by the exhaust gasses to superheat the steam. Note that you can heat steam without increasing the pressure. That thing you were looking at on efunda looks at the line of saturation - saturation pressure and temperature both increase until you reach the critical pressure, which for water, is 3193 psig at a temperature of 705 F.

Not sure what you could do with this superheated steam though. Pressure wouldn't be very high, but you might find some kind of expander (ex: turbine) to recouperate some of that energy. Finding an economical expander might be difficult.

Ivan Seeking
Staff Emeritus