Although the idea of recovering waste heat from a combustion engine is quite baked, I’ve yet to see a system like the one I describe below. It’s relatively simple, lightweight, and wouldn’t cost much to construct, yet it would provide a very substantial gain in energy efficiency.

1) Water is passed through a counter flow heat exchanger, with the radiator input hose, and heated to approximately 90C. This is the preheat stage, and takes heat out of the engine coolant that would otherwise be blown into the atmosphere.

2) The almost-boiling water is used to feed water injectors mounted on the exhaust manifold heat recovery unit. The device the water is injected into resembles a large pipe, sealed at both ends, through which the smaller exhaust pipe passes. It is used to confine the high pressure steam produced when hot water makes contact with the metal of the exhaust pipe and is vaporized. It also acts to serve a constant pressure supply of steam to a small turbine. This pressure is computer regulated through varying the volume/rate of the water injectors, and a variable pressure steam release valve.

3) The turbine converts the steam pressure into mechanical or electrical power (when coupled to an alternator) that can then be used to offload the ICE to some degree.

4) The low temperature steam is then either exhausted into the atmosphere, or partially condensed to supply the repetition of the cycle.

I’m going to make the assumption that the engine I’m using as a heat source is 20% efficient, so there should be 80% of the heat left over to produce steam. At 50% heat transfer efficiency, we’d still have about 40 kW left to run our turbine (with a 100 kW ICE). If the turbine was 50% efficient at extracting energy from the steam, we’d be left with 20% usable energy.

That’s the same amount of power the ICE is making to begin with, so the efficiency of the system as a whole would double.

[Typical gasoline ICE: 20% efficient (after mechanical losses factored in), 650 C exhaust stream heat, 90 C coolant. Makes a great heater!]
-- TIB, Jan 10 2004

internal combustion steam engine http://www.halfbake...on_20steam_20engine
Another heat engine, but this one recovers its water. [phoenix, Oct 04 2004]

Steam Anti Turbo-Lag System http://www.halfbake...0Turbo-Lag_20System
Another, for performance. [phoenix, Oct 04 2004]

Device could aid production of electricity http://web.mit.edu/...ctricitydevice.html
A link from the "internal combustion steam engine" idea. [phoenix, Oct 04 2004, last modified Oct 21 2004]

BMW Turbosteamer http://www.autoweek...002/1024/LATESTNEWS
AutoWeek article about BMW's Turbosteamer [JephSullivan, Apr 10 2006]

Hey [Humanbean], I looked up the Mollier chart, but it was beyond my limited capacity to comprehend :) This idea is based on not much more than the thought that a red-hot glowing exhaust manifold could probably generate quite a bit of steam for real-time energy recovery.

As for the turbine: if vibration and cost are a major issue, then use another type of energy extraction device, like a reciprocating piston. When suggesting the turbine idea, I was hoping that turbocharger technology had brought the price down enough to make it feasible to use a modified version of such.

slight rant: Agreed that Detroit wouldn’t instigate a system like this, even if it worked perfectly, as it’d certainly cut their profits a little. What reason would they have to want to improve fuel economy anyway? It seems the American car manufacturers dictate their rules to the public (and government) and not the other way, otherwise we’d have cars that get 40mpg instead of 21. I’m glad the Japanese will once again kick their ass this decade (Detroit’s), but feel really bad that Americans will continue to lose jobs just to make a few rich bastards even richer.
-- TIB, Jan 10 2004


Good idea! Let's not forget a piston engine is basically a furnace burning (expensive) gasoline and offers a little bit of mechanical output. It is also a 100 year old design - some improvements were made, but still most energy is lost in heat - and a lot extra stuff is needed to keep the temperature under control - why not replace all that stuff and use the heat to convert to energy - for instance like mentioned here.

I'm sure if some effort is put into it (think solutions, not problems) - a major improvement can be achieved. Why are car companies focussing on sqeezing out a last bit of efficiency by electronic control etc - while 70% of the energy is totally lost in heat. I'd say they rather start using that energy!

I'd like to see someone to build a car that uses this and brake energy recovery and see what the overall efficiency will be. Let's not wait for the car manufacturers or oil companies - they are just happy the way things are (until all fuel is exhausted...).
-- inventor, Jan 10 2004


[morrison_rm]: I read about a company, a few years back, that developed a thermocouple generator for trucks, mounted on the vertical exhaust stack. It put out about 1kW of power, which could run most of the electronics. The problem, as always, is cost. Thermocouples are extremely inefficient, and by the time you install enough of them to generate useful power you’d end up spending more than the cost of an alternator. Of course this could be remedied by making thermocouples more efficient (like Borealis is doing), or mass produce them to lower costs. Again, it’s an uphill battle against the powers that be.

[inventor]: It’s true that ICEs are really good atmospheric heaters, that just happen to give off a little mechanical power. I liken them to incandescent light bulbs (another industry/government conspiracy); lots of heat, and a little light to help you feel cozy. I agree that this simple engine modification would be very interesting to build, as it wouldn’t take much effort for someone with the right skills, and I think it’d be fun just as an experiment. I too, would like to see how much energy could be conserved using this system combined with regenerative braking, in the real world. Of course this is just another incremental improvement to an ancient technology (although quite a big increment). The ultimate solution, would be the perfect battery; one that takes in all the power it’s provided, and gives it back with no losses at any amperage. Although this is just a theoretical battery and could never be realized, it is imho, a goal to work towards.

[humanbean]: You make so many good points.

Liability is what almost closed down Cessna. Seems that people flying into mountains has something to do with the design of the aircraft??!?

We are, in truth, trusting more and more of our lives to transistors. Think of old examples like cruise controls, or new ones like automatic defibrillators or the space shuttle. It is an inevitable shift that we must accept. Redundant electronic systems will monitor the state of electric motors and shut down power if needed (starting at the motor, and working back to the batteries).

As for instant acceleration/gratification with a stirling, that’s easy: put it in a series-hybrid configuration (combustion motor > generator > batteries > electric drive motors). When you step on the throttle, there will be instant power provided by a small battery pack, until the stirling spools up. Cost is the only barrier to an efficient Stirling powered automobile, as to achieve an efficiency rivaling a modern diesel would be uneconomical when combined with the cost of the electric drivetrain.

My suggestion doesn’t have to contend with any of the above. It is simple, cheap, user transparent, and in case of an accident poses no more harm than the huge engine block you’d end up hugging.

Is there anyone out there who could calculate how much energy could be extracted from a steam generator like the one I’ve described?
-- TIB, Jan 10 2004


GM tried hard to do a Stirling series HEV. Exotic, high-effieincy prime-movers loose their efficiency benefits when scaled down to car-sized units - too bad. Only a H2 working fluid could do better than IC engines.
-- NatLabgeek, Jan 29 2004


It could work...

The problem is you'd have to carry an awful lot of water. Realistically, condensing the water would require more coolers, weight etc than it's worth... but without them you'd boil the system dry quite fast.

Hmm.

But yeah, it shouldn't be too difficult to implement in a test-stand sort of setup, or even in a car. Sod the powers that be, nobody's actually stopping anyone doing this :-)
-- david_scothern, Mar 15 2004


I once used the exhaust manifold residual heat to warm up my baked bean cans at lunchtime untill one exploded and some beans found its way into the air inlet and the filter clogged. I mean, it was MESSY. Efficiency improved some, though.
-- finflazo, Apr 17 2004


If someone were to attempt this, they should steal the exaust heat from somewhere downstream from the catalytic converter. If I'm not mistaken, the converter requires high exhaust temperatures to work properly. Compromising its performance would cancel out all the wonderful tree-hugging intentions of this idea.
-- JephSullivan, Apr 18 2004


The error here is with that number, 50% efficiency. That is the best that can be had with big turbines using all the advantages of unlimited size weight and complexity, very hot combustion, high pressures and condensing exhaust. None of those are available in a car application.

The actual number for anything that will fit under the hood is about 5%. Half of that 80% waste heat goes out the radiator at temperatures too low to use. If you could collect all of the remaining exhaust heat that would be %40. Multiplying by 5% gives 2% improvement. Even that might be worthwhile but only if the whole apparatus could weight less than 2% of the car. It will work at all only when the engine is running hard enough to produce any exhaust temperature. A turbine specifically is not suitable because any water droplets will erode the blades unless it can be kept superheated, unlikely with such a feeble heat source.
-- Fussass, Apr 18 2004


I recently discovered that a similar system is "in the oven" at BMW. They are creating a drivetrain that recovers heat energy from the engine exhaust by boiling water (and ethanol in a separate system), and converting it into mechanical energy in an "expander". The system is called the "Turbosteamer", which leads me to believe that the expander is a turbine of some sort. The Autoweek article linked above states that BMW claims the Turbosteamer converts more than 80% of the exhaust heat energy into useable power. Also it added 13 hp and 14.8 pounds-feet of torque to the 1.8 liter engine on which it was installed. However, it's not clear that the increase in efficiency and power is worth the added complexity and cost. It will be interesting to see how this turns out.
-- JephSullivan, Apr 10 2006



random, halfbakery