This is from a study NACA (the previous name for NASA) did on water injection a loooong time ago.

The following graph goes like this:

x axis = Fuel/Air ratio (invert it to convert to Air/Fuel)
y axis = Break Mean Effective Pressure (basically torque)

They injected several mixtures. They injected 60% as much of the mixtures as they did fuel. The ignition timing was fixed at 30 degrees BTDC, and the engine is non-intercooled.

Red = No Anti-detonation injection
Blue = Water Only
Green = 70/30 Methanol/Water

Not that with water alone you can lean out the mixture as much as you want and still make more torque (and thus more power) than you could with any air-fuel ratio without Water Injection.

With Meth/Water at this mixture torque could be increased as much as 65%. For comparison, a 65% increase in torque on the STi would yield 495 lb-ft of torque instead of 300. :devil:

http://img.photobucket.com/albums/v14/SaabTuner/NACA812Graph.jpg

Adrian~

I see you are lurking here too, SaabT.

Great graph! Awesome graph!!

It also shows something I've wondered about for ages. What is the optimum AFR.

With water injection it's right around 12.5:1
I've alway assumed that 12.5:1 means 12.5 parts air mass to 1 part fuel mass, in which case the ratio is 0.08. If 12.5:1 means 11.5 parts air to 1 part fuel for a total of 12.5, then it's AFR = 0.09.

Without WI, the AFR of maximum torque is around .11 which corresponds to 9:1, or nine parts air to one part fuel.

A fuel savings of 28%, and a torque gain of 150 units of brake pressure (torque). Incredible!!!!!!!!!

As you know, I'm a biased disciple of WI, so I'll say it before someone else does. If you had intercooling not based on water/MeOH injection, the difference in the curves would be less. But it's hard to argue with such an incredible gain even with a massive intercooler (which has it's own problems).

60% injection is very high as well. I did a calculation based on 50% relative humidity. Interestingly you can inject more fuel than water and still stay under the dew point for air at sea level at 75F. I did this calculation because I thought that even at 25% water:fuel injection it might saturate the air and cause all sorts of problems.

Curious SaabT, is this direct WI (into the cylinder), at the throttle body, or somewhere up stream. Is it a turbocharged engine?

Also, with the MeOH, it's hard not to discount the energy added by the methanol, which is effectively fuel. Even though the stored energy in MeOH is less, it is still considerable. This addition could effect both power and the AFR peak. However, with water alone the AFR peak (plateau) is also in the 12.5:1 range.

That test was at a fixed air injection temp of 250F ... here's a graph at 150F (which I should have shown first, as it's more indicative of an intercooled engine):

http://www.imagehdd.com/d1y2004/3187NACA821_Intercooled.jpg

The engine is fed pre-compressed air from a controlled source. The coolant temp is fixed, the compression ratio is 7:1, and the ignition timing is at a fixed 30 degrees BTDC.

The water was injected just prior to the inlet valve.

Adrian~

As you can see in the second graph, which would be like an intercooled car, while the difference between the curves is less, it's not very much less, and in some places more.

Needless to say a 70/30 mixture of Methanol/Water making over a 50% increase in power even at 14.7:1 or even LEANER, is incredible.

With internal coolants, you can run as lean as 25:1 and still produce more torque and power than just gasoline alone at 12:1.

Adrian~

Some years ago there was a company that sold a Water Injection System for v8's that operated from the exhaust manifold pressure, this same principle would probably work on turbocharged engines by allowing Boost pressure to determine when and how much water to add. I believe the WRC cars use Water Injection too. Seems like the perfect solution for California 91 Octane Gas, and anyone else looking for a way to run more boost and timing without blowing up your engine. This would solve the boost creep issue with STi's, the higher the boost, the more water sprayed, no detonation.

.
:lol:

I plan on rigging one up which is basically a duplicate of the fuel injection system, but with slightly smaller injectors. Then I just run an auxilliary injector driver, and put a 5 gallon tank in the trunk. (No worse than some of the Subwoofer boxes I see in some cars.)

It'd be economical, power-producing, and badass all at once. :lol: And it'd be unique, which is often enough in itself.

Oh yeah, and CARB allows water injection, whereas they do NOT allow exhausts, intakes, bigger injectors, etc etc ...

Water injection and intercoolers are probably the only two modifications that CARB doesn't give a rat's ass about. There are a few other things you can do as well, cat-backs and such. But almost every other mod is illegal. To get CARB Executive Orders you have to prove that your mod did not produce even one ounce more pollution than stock. With a WRX, that's about impossible because you'll HAVE to remove that up-pipe catalyst ... and there's no way you can remove that and produce LESS smog. Not gonna happen .... even if you could, CARB wouldn't allow it because during warm-up without a catalyst there, you'd still pollute slightly more. Oh well.

Adrian~

Ahh yes.... CARB, the reason I will NEVER reside in the wonderful mudslide, earthquake, forest fire'd, rolling power outaged state of California! ;)

P.S. great write up Adrian :p

Jorge (RiftsWRX)
www.ProjectWRX.com

Another great graph, this one we can see the effect of reduced temp. I know it's not a perfect mimic of our engines, but I'd say it reflects what Wi can do.

So, at low enough temperature, you can run leaner than stoichiometric, and still produce as much power as 12.5:1 or even 9:1. It makes me think about the idea that the reactants are preventing complete combustion, unless there is excess oxygen. Obviously, these data would support the idea that IC engines are knock limited with respect to fuel economy, not some thermodynamic air:fuel magic ratio. That ratio, would seem to be a product of combustion chamber itself, the octane being used, and cylinder temp. Quoted as dogma, it is clearly not necessarily optimal in all engines--something I think we should all think about.

I know SAABT that you have an interest in lean burn engines, and these data would seem to support no downside in power given adequate cooling of the air. I must admit a certain fear that lean burn combustion is just another layer to the house of cards being built around water injection as the water permits operation outside of normal fuel ratios. The risk of detonation increases even more under lean burn conditions than when tuning to 12.5+WI.

It seems that total confidence in WI is what is needed, and this means constant monitoring, short term back up systems, and computer control over shutting down fuel or blowing off boost in the even of a clog or failure.

Similar worries must have been present when fuel injection was new, blockage or failure of an fuel injector could lead to the same lean detonation favorable condition, yet we don't worry about fuel injectors now.

The Aquamist DDS2 system offers some relief by monitoring flow and the option of signalling any of numerous devices of a problem. Additionally, I'd like to see a system that provides water even in the event pump failure, high speed valve failure, or development of a clog in one jet and not another. More monitoring at each possible point of failure, and a system of compensation for any failure would go along way to reassuring WI users.

One additional means of off-setting a pump failure is to equip the system with an "accumulator" that will sustain pump pressure and flow even after a complete pump meltdown. Development of a 'back up' jet to be used in the even of a component failure, that only injected water in the event of a failure would also make a more fail-safe system (perhaps using boost pressure as a source of water flow). In fact, "over injection", that is injecting lots of water (flooding), would be an excellent response to any failure, one that would safely indicate a problem (power loss) and also protect the engine from detonation.

It is a excellent alternative to drowning the engine with fuel just to cure detonation. I think you are fighting a loosing battle - most people nowadays preferred running rich a/f ratio - including tuners.

Good luck

You may be interested to view this data logged chart on the effect of water injection. Thermocouples wer mounted: pre-turbo, post turbo, pre IC and post IC. Temperature at the manifold had dropped below ambient ! on a relatively cold day.

http://www.aquamist.co.uk/phpBB2/viewtopic.php?t=514