Meyer Micro Energy Lab
1-JUN-2001 updated 6-JUN-2001
Water Injection Experiments in my 1993 Geo Metro
Water injection into an internal combustion engine is supposed to increase power and gas mileage as well as moderate combustion temperature and even clean the comustion chamber, etc...

Well, I tried this about 20 years ago on my 1975 Buick LeSabre and it didn't work too good, at least the way I did it didn't!
Back then all I did was spray water into the carb. with a windshield washer pump, tubing and 2 needle-valved venturi's from model airplane engines, placed over the primary barrels of the quadrajet.

So, why am I trying it again?
Well, I think I have a better way of doing it this time, that should work.
After learning more about the combustion process in the engine and looking at other setups on the 'net.

So, this is the setup that I think will work now:

The water tank (A) has a vent tube (C) which connects to the exhaust manifold (I) and a water tube (B) that connects to the heating tube (D). The heating tube (D) is wrapped around the lower exhaust manifold (I), it connects to bottom of the boiler (E). Out of the top of the boiler is a tube to the needle-valve (F) then more tubing to a 'tee' fitting on an existing vacuum line (G), connecting it to the vacuum port in the throttle body (H), above the throttle plate.

See Under-the-Hood Pictures

A - Water tank

B - Water line from tank

C - Pressure/Vent line to tank

D - Heating tube

E - Boiler / Liquid-Gas seperation

F - Needle Valve

G - Existing vacuum line

H - Throttle Body

I - Exhaust Manifold

H2O-Injection Diagram

How does it work?

Water is moved thru the system by the combined action of the vacuum at the throttle body and exhaust gas pressure from the manifold. This 'balancing act' varies with driving conditions, modulating the amount of water to the engine according to need.

When water is heated in (D) and starts to boil, expansion of steam pushes back water towards the tank, providing some self regulation of water flow into the hot section, bleeding excess pressure back out the vent tube into the exhaust manifold and moving water/steam to the boiler (E) where it is further heated and space for seperation of liquid and gas can occur so that only steam is output thru the needle valve (F) being drawn into the throttle body (H).

Stay tuned for some more results... Notes:
First trials
, with no valve (F) or boiler (E), soon showed that it was easy to get too much water moving thru the system with the result of lots of liquid water going into the engine thus degrading performance - yeah, like hardly running, missing like crazy... like; will I make it back to the driveway! First drive used over 1 qt of water in 2 miles. Crimping the copper output tube helped until the valve was installed.  I drove it to work (7 miles), was a little rough in-town. When I started it to go home, the valves were hung up at first - "Oh Crap! What Have I done!" On second try it started up... too much water yet!
Next came the valve(F), only have it open 1/8 turn or less. And can shut it off - goody!  Still too much water - liquid not good, need steam!

Then the boiler/ seperator thing(E). Drove it last night and today, much better! Filled up the gas tank, time to see what kind of mileage I get... more on that when the results are in.  

In the engine:
Here's what I think goes on in the engine, at least some of it... 

When the steam enters the throttle body and mixes with the air/fuel mixture it chills, condensing into micro-fine droplets of water, it also gives some heat energy to the air/fuel helping more fuel to vaporize on the way into the cylinder - these are good things. When this mixture is compressed and ignited the temperature and pressure rise alot, very fast, the water droplets absorb heat and change state to a gas (steam) - this moderates the combustion, another good thing. The steam expands over 1500 times its volume increasing the pressure on the piston thus giving more power - a very good thing! - unless your cylinder head is loose!

Observations so far:

Noticed increase in power - less pedal / more zoom! Seemed to have more pulling power, engine was 'eager' to go...
Also, the engine sounded quieter, smoother when H2O injection on, yet more power!

Next week I'll be going to Pontiac, MI - over 200 miles each way, so I'll get some road testing and mileage figures for sure!
Think I can I get 70 mpg???
My Car:
1993 Geo Metro, 4-door hatchback, 5-speed stick, 993 cc engine displacement, 3-cylinder, 52hp / 5700 rpm, 58 ft lbs torque / 3300 rpm, 206,000+ miles, getting a little rusty around the edges...
Mileage: (without H2O injection)
Best trip
on 20-MAY-2001, Lansing to Shelby, 152 miles on 2.47 gallons of gasoline = 61.5 mpg! Highway driving at about  63mph (100 kph). This also utilized an "EFIE" to lean the engine a bit.
The next best trip,
May of 2000, same trip, same speed, similar weather, no "EFIE", was 52.5 mpg.  
From Eagle Research:
The EFIE (Electronic Fuel Injection Enhancer) adjusts the voltage level from the O2 sensor to the car's computer to lean the engine's fuel/air ratio.
EFIE web link.

See the Under-the-Hood Pictures

Back to Energy Lab Contents