I’ve brought this up before but it didn’t seem to garner much interest. I am hopeful that with oil at $100/barrel and gasoline well over $3/gallon that this idea might now take hold because it has the potential to significantly improve fuel economy, power, and reduce emissions. This is something I feel fairly passionate about because I’ve done it! It’s not a hypothetical for me, I’ve modified a vehicle and put quite a few miles on it.

Let me start by stating my practical experience. I put a water injection system on a 1968 Rambler American with 199 cubic inch V6 engine. This car was a mid-sized passenger car weighing approximately 3200 lbs.

Without water injection, the engine really did not have adequate torque for how the car was geared (2.7:1 rear end and three speed manual transmission). It was picky with respect to fuel and knocked under heavy load. The fuel mileage without water injection was approximately 16 MPG city, and 22-23 highway. Tuneups were necessary every 10,000 miles or so and the spark plugs would be significantly dirty by that time.

I added the Edelbrock water injection system. The Edelbrock is flawed in major ways. The only really good aspect of the Edelbrock is the adjustable controller that can adjust water volume according to both vacuum and RPM. The problems with the Edelbrock are that the tank is far too small and the pump is integral to the tank making it impossible to just replace the tank, and, there is no filtration system so the injector rapidly becomes clogged.

The literature recommended about 1/10th the amount of water to fuel; however, I found optimal performance with city driving happened with approximately as much water as fuel, which made the 1 gallon tank useless. On the freeway when the engine was not so loaded, 1:10 ratio was closer to, but not quite, adequate.

The mileage improvements were not realized with just the addition of the water injection system, but with adjustments made possible by it’s addition. Stock, the engine would ping under load with the factory specified 6 degrees before top dead center initial advance and with the factory carburetor jets. The water injection totally eliminated this ping.

I found it possible to increase the initial advance to 14 degrees with no pinging and a great improvement in power across the entire RPM range. Particularly, the lack of pinging improved the usable low end torque significantly making the car a lot more pleasant to drive.

The 14 degrees initial advance and water injection both had the net result of significantly increasing the idle speed, so I was able to cut down on the idle settings quite a lot and save fuel. Also, no longer relying on fuel to cool the peak cylinder temperature, I was able to jet the carburetor a couple of jets leaner, as well as lean out the idle mixture, without pinging, and still obtain significantly more horsepower than stock.

The limit to how lean I could idle it was where it would start to miss. I added an MSD multistrike ignition system and that extended how lean I could set the idle without missing.

With these modifications, I now got 30 MPG on the highway, around 23 MPG around the city (I could stretch that to 25 if I really grannied it). What’s more, this car had no catalytic converter, but I had some tests done on it to see how clean the engine was and it came well within current smog requirements, far cleaner in fact than a brand new catalytic converter equipped GMC van we compared it to. An additional benefit is that it almost totally eliminated the need for tune-ups. Still had points but since they switched no current with the MSD ignition, they lasted; and after ten thousand miles I’d pull the plugs out and they’d look like I just pulled them out of the box.

Since much of what wears an engine is acids that are formed when incomplete combustion products react with water vapor, the almost total elimination of those products also had to be good for engine wear. Unfortunately, there were other aspects of the car that were mechanically bad, in particular the three speed no synchronizing gear in first transmission and completely inadequate brakes (at least for the way I drive), that ultimately lead me to sell it before the engine could approach true old age.

I’m convinced that even greater fuel economy and power could have been obtained by raising the compression ratio since there was absolutely no sign of pinging even on the worst obtainable gasoline.

These modifications approximately doubled freeway economy and improved city driving economy by more than 50%, while at the same time improving the power and response of the vehicle and cutting pollution at a cost far less than hybrid modifications. However, I see absolutely no reason this technology could not be combined with hybrid technology to gain the additional benefits provided by hybrid technology and such a combination might easily yield full sized drivable vehicles with fuel economies exceeding most motorcycles.

A factory equipped car would have two similarly sized tanks, one for water and one for fuel, and since the economy would be much higher, the fuel tank could be approximately half the present size so really no additional space would be required in total for the water storage, it would simply displace some of the fuel storage.

In a conventional engine, much of the heat energy is absorbed by the cylinder walls and thus not available to provide propulsion. In addition, the higher cylinder wall temperature reduces the overall temperature difference from fully compressed to fully expanded, and thus engine efficiency.

Water injection provides droplets of water vapor that are vaporized into steam by the flame. This simultaneously cools the flame while turning into steam and increasing in volume providing greater push to the piston. Less heat is lost to the cylinder walls.

Water injection allows a leaner mixture which would otherwise result in excessive cylinder temperatures, pinging, and engine damage. This in turn provides better economy as well, and reduced in carbon monoxide emissions. I also found the hydrocarbon emissions to be greatly reduced, in fact both hydrocarbons and carbon monoxide were almost unmeasurable. Nitrous oxides were reduced enormously as well, and the need for EGR eliminated as the water vapor provided all the positive benefits of EGR with none of the negatives side effects (reduction of power, etc).

If American car manufacturers weren’t such dolts they’d catch on to the fact that this is an inexpensive technology that would allow them to produce a car that could compete with the Japanese hybrids for economy, efficiency, and pollution specifications, while at the same time being much less expensive and having no battery pack to wear out.

Alternatively, they could combine this with hybrid technology and produce vehicles that were more efficient, more powerful, and lower in emissions than the Japanese hybrids at a similar cost.

Other possibilities also occur to me; for example, there is a class of carburetors referred to as “supercarbs” that various experimenters have built that use engine vacuum to vaporize fuel and provide replacement heat from radiator water, etc.

These carburetors can provide extreme improvements in efficiency because instead of feeding gasoline as droplets which then take energy from the flame to fully vaporize as they combust, feed it as a vapor. This results in slightly better combustion but the main efficiency gain comes from the fact that energy to vaporize the fuel is drawn from radiator waste heat not from the flame.

These super carburetors have not been practical for the most part because such a totally vaporized mixture is prone to pinging, and so far the inventors have been largely unsuccessful at working out metering and mixture issues over the full RPM and power range of the engines. None the less, I’ve seen some incredible mileage result from these. Water injection would at least eliminate the pinging issue and while it might be a stretch for the home experimenter, a modern closed fuel cycle approach applied to supercarb technology, combined with water injection, might yield efficiencies greater than fuel injection.

American car manufacturers are dolts, or at least run by them, so I’m sure they’ll continue to try to produce and sell inefficient SUVs while gas is $3+ a gallon and wonder why their losing billions of dollars a quarter.

Edelbrock no longer makes the water injection unit but there are other companies that do, such as “FJO” and AquaThrust. They tend to be more pricey than Edelbrock was (Edelbrock unit cost under $300 while the AquaThrust is almost $800). If you do decide to do this pick a unit with a separate pump, add a filter inline before the injector, and use a tank approximating your gas tanks capacity.

FJO Water Injection:

Aquamist Water Injection System:

AquaTune Water Injection Systems: