water induction + wai advice?

[lunarhighway]

i want to make a simple water injection or rather induction setup, to see if there's any gains to be had. i think in combination with my curent wai setup that makes for bigger throttle settings and a leaner mixture there's several things attributed to water injection that could benefit me, even it gains are small i dont mind refilling a water tank every day as water is free and fuel is not, plus i could always fit a bigger water tank.

the basic idea is to make the intake air more humid, but i'm wondering what the best way would be to do this.. i've looked into several water injection systems described on line, but i'm wondering what would work best.

would the lower pressure in the airbox, at wide throttle settings, be enough to suck in water from a reservoir trough a fine needle? alternatively in stead of sucking in water directly i could aslo employ a "bubbler" system where air is sucked from the top of a closed water reservoir and this air is replaced with air that enters trough a submerged intake, thus creating bubbles making the air that's sucked into the airbox take small water doplets with it.

has anyone ever tried something similar or can anyone offer some advice?

[Daox]

FYI your warm air intake doesn't lean anything out. Your O2 sensor corrects the less dense air with less fuel so you retain a stoichiometric air/fuel ratio.

That being said, when I was looking into water injection (a while back, not sure it has much merit anymore to be honest), I wanted to use a fuel injector to atomize the water as much as possible. You could easily tap into an existing fuel injector signal to meter your fuel very precisely according to the amount of fuel the engine is using. Most commercial kits are crap imo.

[markweatherill]

I don't have the answer but I would like to know if a little electric nebulizer would do any good in making intake air humid? They produce a very fine mist, for sure.

[mpgx2]

Foggers aren't so good for WI because they don't
make enough mist.

Daox, I'm so impressed with your FE. I bought a
Alltrac celica just recently and am hopelessly
unimpressed with the FE. They should just call
it a mini V8 landcruiser - lol.

Yes, most of the commercial WI systems are
low-tech. I'm certain that an arduino solution
could be much better.

Apparently, it's possible to replace up to 25% of
fuel by mass with water.

[procupine14]

Originally Posted by mpgx2

Apparently, it's possible to replace up to 25% of
fuel by mass with water.

My only concern with that would be needing to be very careful to not hydrolock the engine if you get too close to that mark

[Daox]

I don't think you'd ever want nearly that much water injected into the engine. In fact, I don't think you want to inject much at all. If the point is to create steam from the water so the expansion of the steam aids in powering the engine, you want to inject already hot water (easy enough, just use coolant to heat). But, you don't want to inject a ton because the phase change takes quite a bit of heat energy to complete. You might end up just cooling down the combustion, not turning much water to steam, and thus loosing power and decreasing efficiency.

[lunarhighway]

I don't think you'd ever want nearly that much water injected into the engine. In fact, I don't think you want to inject much at all. If the point is to create steam from the water so the expansion of the steam aids in powering the engine, you want to inject already hot water (easy enough, just use coolant to heat). But, you don't want to inject a ton because the phase change takes quite a bit of heat energy to complete. You might end up just cooling down the combustion, not turning much water to steam, and thus loosing power and decreasing efficiency.

i certainly wouldn't want to get a whole lot of water in the intake, i'd prefer something comparable with what it would get driving in rainy conditions with loads of spray.

my dad once experimented with steam when he was young by trying to boil water using exhaust heat. unfortunately, they never got enough heat to get propper steam.

i'm thinking about routing a pipe to my airbox trough my wai and poking an injection needle trough the air filter, on the other hand perhaps connect this pipe to the top of an air reservoir and have the reservoir such in air trough a submerged bubbler. this way it would such up suspended air bubbles wich would than pass trough a warm duct and be sucked up above the air filter where i imagine is more low pressure than below it.

but perhaps if u use a bubbler i need a bigger intake than a syringe needle?

[busypaws]

Here in Tucson right now we have 4-8% humidity and over 100F). My Toyota starts pinging at 95F with this humidity. I've taken a small plastic soda bottle and strapped it to the stearing column. Then run a drip irrigation hose from the bottom into the top of the air filter. (Gravity fed - Engine side of filter). Here is put a dripper set to drip about two drops a second. So after a minute or two my filter element is wet and I'm getting cooler and moister air into the engine and my pinging problem has gone away.

It's a Swamp Cooler for my air intake.

Might not work if you already have >50% humidity. But here in the desert increasing the humidty from 4% to something approaching 50% keeps me from having to run Premium fuel in the summer.

[mpgx2]

Another toyota man with a heat engine - lol

That's a novel idea using the air filter and wetting it.

Apart from more power, are you able to put a number on
the improved FE that it gives? So I take it you are leaned
out somewhat from the fact that you get pinging?

Some Vendors of WI systems claim 20-50% improvements.

The 25% water by mass comes from a racing site. Obviously
if the motor is hotter then it seems logical that you could
use more water than if it is cold. Its hot here too that's
why I'm attracted to WI.

[gone-ot]

...Swamp Cooler Theory 101:

eqn 1: E(s) = ( DBD / WBD ) = (T1 - T2) / (T1 - T3)

...where:
E(s) = Saturation Efficiency, actual change divided by maximum possible change.
DBD = Dry-Bulb Depression, air temperature change from HOT to COOL, actual.
WBD = Wet-Bulb Depression, air temperature change, HOT to COOL, maximum possible.
T1 = Outside (or inlet) HOT air temperature.
T2 = Inside (or discharge) COOL air temperature.
T3 = Wetbulb temperature, maximum adiabetic air cooling possible.

...for example, if T1 = 105°F and T2 = 73.9°F and T3 = 66.1°F, then the saturation efficiency E(s) would be 0.80, or 80%:

E(s) = (105.0-73.9)/(105.0-66.1) = 0.80 x 100% = 80%

...unfortunately, that E(s) value is for Aspen Excelsior, not a FRAM™ paper automotive air filter, which probably has an E(s)-value of ~ 0.5 or less.

...never the less, let's backsolve the above equation to see how "cool" the air going into the engine (ie: T2) would be with E(s) = 0.5:

eqn 2: T2 = T1 - E(s)*(T1 - T3) = 105.0 - 0.5*(105.0 - 66.1) = (105.0 - 19.4) = 85.6°F ...a (theoretical) temperature reduction of 19.4°F.

...if you know the relative humidity (RH%) values of T1, T2 and T3, Eqn 1 becomes:

eqn 3: E(s) = SQRT[ (rh2 - rh1)/(rh3 - rh1) ]

...where:
rh1 = relative humidity at T1 of the inlet (outside) air.
rh2 = relative humidity at T2 of the discharge (after the wet filter) air.
rh3 = relative humidity at T3; theoretically 100%, but practically less than 80%.

...so, because of the HUGE volume of air drawn by the engine through the air filter, a large amount of water is required to raise rh2. For instance, in the first example above, the outside air would be: T1 = 105°F @ rh1 = 12% and the inside air (going into engine) would be: T2 = 73.9°F @ rh2 = 68%.

...unfortunately, THAT's gonna take a LOT of water, especially if you're driving at freeway speeds (higher RPMs) for any appreciable time...literally, gallons-per-hour!

...and, here's the equation for calculating "how many" gallons-per-hour (GPH) of water you'd need:

eqn 4: GPH = [CFM*WBD*E(s)] / 8700

...where, 8700 is a conversion factor based upon (a) 8.34 lb. water-per-gallon and (b) 1043 BTU-per-pound of water. CFM is a function of engine displacement (CID), volumetric efficiency (~ 80%) and RPM:

eqn 5: CFM = (CID*RPM*0.8) / 3456