Toyota Forum banner

The Free Mod Series: Part VII Cheap Water Injection

1 reading
13K views 17 replies 8 participants last post by  73sport  
#1 · (Edited)
Background:

Pre-Ignition, Detonation, Knocking, Pinging along with Dirty, Grimy, Ass-tastic carbon deposits. (which heat up tremendously and also cause pre-ignition) It's hard to put together two things that will hamper performance, and destroy a running engine in a shorter span of time!

There must be a way to easily defeat them both, without compromising performance, and reliability.

Theory:

Enter water & water/alcohol injection! We'll skip technical detail for a short list.
Water injection suppresses detonation like no other. It accomplishes this in three basic ways:
1) The incoming air charger is significantly cooled by the atomized water
2) Combustion chamber temperature, if a larger amount of water is injected, are lowered.
3) The water cleans off carbon in the intake track, while the steam created in the combustion chambers blasts carbon off parts clear into some of the exhaust manifold.
Because of this, depending on the amount of water / mixture injected, the octane rating of the fuel is effectively raised 15-30 points!
How much water is needed? Well. That is a loaded question. Typically, you add your mixture, which is now both fuel AND WATER and take 12-25% of that.


Water/alcohol injection.

Short and sweet! Adding up to a 50% alcohol ratio to the water will increase it's charge cooling efficiency. We do *NOT* NOT NOT NOT want to injection alcohol as an extra "fuel". IT DOES NOT WORK THIS WAY!!!

There are three types of alcohol you can inject: Methyl, Ethyl, and Isopropyl.
Isopropyl (rubbing alcohol) CAN NOT BE USED!!! It *will* try to burn during the combustion process and suck a significant amount of oxygen from the gasoline.

Both Methyl and Ethyl WILL NOT burn in a majority gas combustion process... It's next to impossible as they are under their lower explosive limit (the ratio of air to fuel needed to combust). Isopropyl WILL try to burn, as it is just above it's lowest explosive limit. That causes the mixture to be extremely rich.

That leaves us with methyl, or ethyl alcohol. Throw out methyl! It eats Aluminum over time!!! That leaves us with regular ethyl alcohol. It can be bought in paint stores for $10 a gallon.


Instructions:

Sadly, you can't just go out and buy fuel injection parts. Water conducts electricity, and the Alcohol will eat away at gasoline safe parts. You would corrode a fuel injector shut rather shortly. Noone wants to rebuild an injector every month :\.

Thankfully! Toyota has given us a gasoline injector that *CAN* tolerate water! The old style Cold-Start Injectors!!! They consist of nothing more than an atomizer, and a solenoid to open/close the injector.

Now we have two upsides, and a downside. On the good side, unlike just buying an atomizer nozzle, when the injector is turned off, boost will NOT flow down the injector/water will NOT be sucked out by vacuum. This is a simple injector, NO DRIVING equipment is needed!!! That's also the bad news, being a simple injector, it's either on, or off with nothing in between. If you don't like it, you can simply cut the shaft off and use it's atomizer.

What does a used cold-start injector cost??? Try $1.50 on ebay. Less than $5 shipped! (2vz-fe / 3vz-fe CSI's are fine. The 2vz-fe injetor is 1 1 /2" longer, so I would take a 3vz-fe injector before that as it would have an easier time fitting in the middle of the intake.
The flow rate for a clean injector is around 150cc/min (OF GAS, NOT WATER) at 30psi.

What do we do for a water supply? Seeing how engine bay space is at a premium, and we all-ready have a 5 quart water supply installed; al-la washer tank. That's the supply!

Water pump. originally I bought a windshield washer pump. (again, used Toyota part shipped for less than $5 on ebay) Unfortunately, this does not deliver enough flow for a boosted setup on it's own. I couldn't use it. I am sure if you bumped it's voltage up to 18-24v it would be a *perfect* pump to use!

So... I was forced to drop a spare gas pump I have laying around! All 80lph of it. It's hard to find information either way about submerging a fuel pump in water, and pumping it. Then again it's a spare so I don't care what happens.




I have simply hooked both the pump, and injector up to a 40 amp relay (again, bought on ebay for $2.50, shipped for $4). Which is triggered by the switchable ground on my SMT-6. This way I can trigger it by any combination of the following: Throttle position, RPM, A 0-5v source, another 5v source. (such as, a map/boost sensor, and temperature sensor)

For anyone *not* using an SMT, you could do something like using a boost switch. They're typically found in 1-4psi, 5-14psi, 14psi+ versions for $15-$25 new. This isn't the best way to do this as the water would be trigger many times in normal driving when it shouldn't.



Originally, the flow was simply insane! The water pressure in the line must have been phenomenal as the 150cc (gasoline) CSI was flowing somewhere around 600cc/m !? That's the flow you would want around 450-550hp on a boosted engine!!! It could empty a one gallon milk jug in just over 4 min!

Something had to be done! I used a smaller line to restrict water flow, and a smaller line to restrict water intake. Now we're down around 450cc/m as It takes just over 60 seconds to empty a 20oz coke bottle!

Eventually, I will wire a large potentiometer to lower the output further at a later time.






This particular digi-cam has a hard time picking out the water, since it's installed, I can't take it out and put it against a bright backlight on a black background. So we'll make do!




Image



Image



Image



Image



Image



Image








****************************************

One other thing worth mentioning... When you're on boost running water injection, you have no reason to fuel past your peak power. (typically 12.5) The water is doing all the leg work of cooling the charge, and much better at controlling the pre-ignition in the combustion chamber than vaporizing gas is!

****************************************


****************************************
Other Articles in the Series:
Part I Quicker Transmission Shifting
Part II Adding Airflow
Part III Taking Advantage of Intake/Exhaust Upgrades
Part IV Cheaply Monitoring the ECU & A/F
Part V How to pick up MPG, and sleep with her too!
Part VI Acing Aerodynamics
Part VII Cheap Water Injection
Part VIII Electronic Transmission Controller
****************************************
 
#2 · (Edited)
This is mainly for Forced Induction guys, but...

A washer pump would create a *correct* ratio of water to mixture mass on a near stock N/A engine (say a sub 280hp N/A v6). I've noted that the engine runs much smoother while on the water. It also stays clean so it's fairly well worth the $15-$30!


Now... There is *nothing* that says you can't do something silly like inject nitromethane and methanol into the engine. Just ask my knock sensor I just had to fix!:lol:
(Good for another 10-15hp! :clap::)
 
#5 · (Edited)
Ty. Even after fixing my knock sensor problem, I've still got some massive knock at just over 4000rpm. Took it out and flogged it this afternoon (with no exhaust, aint Alabama grand!) Wouldn't do it under any condition I tried. So it definately works! :p (I'm happy it was detonation! I was expecting a failing bearing, or some such trickery)

I just wish it was varriable... If it was, it'd be the perfect solution. Set it up low on the scale. That way it's not dumping way too much water in at lower power levels!

I just had mine switching based on TPS, and over 1500rpm today. Lemmie tell ya... The engine doesn't like 300cc of fuel mixing with over 400cc of water hahahaha! It kinda had a studder till I set the minimum RPM to 2500.

My detonation also helped show something else. This setup, from an un-filled hose will be squirting water in around a half second. Can't take any more than 1500rpm for the thing to come online, so that's great.





I'm still concerned with flow rate on this fuel pump. I want to T a smal diameter return line at the injector to help cut flow rates more. The line is bad to blow off too. I had to zip tie and hose clamp them together, and they still weap water. Atleast they don't blow off anymore!
 
#8 · (Edited)
I just drilled a hole smaller than the CSI into the stock rubber intake hose. Right behind the air meter.

Shoved it so the nozzle would be around the middle of the pipe and left it unsupported. Dripped some water on it with the engine idling and there was no leak so... That seemed to work well enough.



ekim55121 It's fairly simple.
You need hosing, wire, a CSI, a pump, a relay, and something to trigger the relay with.


In this picture, Red is power. Black is any avalible ground. (Engine, or chassie)
The air-switch (vacuum for N/A, boost sensative for boost) activates the relay. The relay trips (a relay is nothing but an electriclly operated switch. Usefull in having a small electrical current, activate a much larger electrical current), allowing power to flow to the pump and CSI. The pump sucks water from the tank, sends it to the open CSI. The bypass is optional, depending on how much flow you have.


Image



A Toyota windshield washer pump flows enough for a turbocharged, stock i4. You would blow the stock i4 just before the water injection started getting out of it's msot effective ratio.
 
#10 · (Edited)
so you are saying if i do this to my motor and its non turbo charged it will sitting next to my supra in the driveway? also waiting for a new motor?
I'm sorry ekim but this doesn't make any since. Would you re-word it for me?


Pete; I know exactly what you've been trying to do, and It's just not gonna work AFA I can see! ;) You can try a humidifer, but that's not much water. I was seirous, that digicam sucks and you can't see the amount of water spraying. It's an extremely fine mist, and a lot of it.



Pete do this. Take whatever toyota CSI you can get your hands on and shove it into the intake (remember it *must* be behind the airflow sensor!!!) Water will kill a MAF, and on a vane-flap, any fluid is going to trigger an Intake Air Tempature malfunction code.


Take the CSI and ground one wire. Take the power wire and splice it into the stock washer pump wire. Pull the washer pump hose off, and put a new line going to the CSI. Now all you need is a way to trigger it. If you're not afraid of rain then it's easy. You simply pull the electrical feed to the wiper engine. You could also just cut the wire that triggers the wiper motor, and put a switch in it. That way you can disable it.



Probably cost $5 to do it that way.
 
#13 · (Edited)
Pete you'll find this interesting:

national Advosory Committee for Aeronautics
1942

Investigation of water induction in a single-cylinder engine over a range of fuel-air ratios from 20:1 - 8:1 indicated the following conclusions:

1) Water injection allowed a fuel to be operated above it's normal maximum permissible performance limits.
2) Water injection allowed a fuel to be operated at a higher indicated mean effective pressure, with a lower indicated specific fuel consumption, or with both, than was permitted without an internal coolant.
3)Water injection had a marked cooling effect on the engine head an cylinder. The exhaust-valve-guide was the only point on the head at which the temperature showed a tendency to increase with indicated mean effective pressure. The temperature was less, however, than that obtained with a straight fuel permitting equivalent power.
4) Water injection showed no advantage in fuel economy when the fuel was operated well below it's maximum permissible performance limits.


Based on that, Without re-tuning the ECU with more ignition advance, and a leaner mixture, I wouldn't think the fuel economy will improve much. If it did, I would expect that to be because carbon deposits are being cleaned away.


********************

Sean, David, Tony, and Justin - yall will find this interesting!

All the testing was done on a 202ci single cylinder air-cooled engine. 7.0 Compression ratio, 80 octane fuel, and the engine is being force fed air through it's manifold. Keep in mind this is all old-school stuff, so don't be trying to compair that to modern stuff. Taken at a manifold inlet tempature aof 250*F, at 2000rpm.
(The testing methodogy was to increase power until the engine was audiably knocking, then back off from it's onset 7%)

Indicated mean effective pressure = combustion chamber pressure
Maximum Manifold pressre = boost in in/Hg scale
Estimated hp = you're a retard if you don't understand this one!

At an Air/Mixture of 15-1
100% fuel:
IMEP = 180psi
Boost = 37in/Hg
Est hp = 280

20% water to 80% fuel:
IMEP = 215psi
Boost = 43in/Hg
Est HP = 330

60% water to 40% fuel: ***
IMEP = 306psi
Boost = 59in/Hg
Est HP = >450***

That's at a freaking 15-1 Air to mixture ratio!!!


At a more friendly 12.5-1 ratio
100% fuel:
IMEP = 205psi
Boost = 42in/Hg
Est HP = 306

20% water to 80% fuel:
IMEP = 264psi
Boost = 52.5in/Hg
Est HP = 390

40% water to 60% fuel: ***
IMEP = >310psi***
Boost = 62.5in/Hg
Est HP = >470***

At 10.5-1 ratio
100% fuel:
IMEP = 260psi
Boost = 52.5in/Hg
Est HP = 385

20% water to 80% fuel:
IMEP = 295psi
Boost = 58in/Hg
Est HP = 425



*** Results not exactly measured. Engine limits discovered.



I included the 40% mixture purely to show just *how much* fuel is having to cool the mixture, and that yes you CAN lean fuel safely, provided you give thought to the amount of water needed to offset the fuel.
 
#14 ·
As good as this sounds, most guys will not find any benefits from using water injection on a stock motor with stock engine management :lol:

I know it was a good way to get rid of your detonation problems, but I am sure that a couple of degrees of ignition retard will solve that.

Here's a good thread about water injection... I also contributed in this discussion as well:
http://www.honda-tech.com/zerothread?id=1208361&page=2
 
#15 ·
I'm not advocating it to stock people, just throwing it out there. (I've said multiple times the main benifits are nil if you're not really on the verge). pgmomni has been wanting to try water injection on his stock engine for a long time.





Anyways. This will help me with the turbo, when everthe F'ing MIG get's off back order and get's shipped. I just figured someone might like some info on it. If not until a long time in the future.
 
#17 ·
interesting note here.

i read this and got motivated to do this to the scepter, and found a few parts that might be a little different

i found a csi for a 79 280z is about the right length for 2.5 intercooler piping, and, im using a 92 jeep cherokee washer tank. it has 2 pumps, and a level switch. i was thinking high and low with a dummy light. ill get some pics up when its installed