DIY water alcohol injection setup!
 

I did a water/alcohol injection setup over the weekend with great results. Since I still dont have the manifold I want and a regular intercooler, some chemical intercooling was in order. I'm planning to take it to the track this week and its the last test and tune before the season is over. I had to do something!

The nice thing about this setup is that its boost sensative, as in the more boost you make the more it will flow. As boost drops off the flow does too, so your not stuck with a on/off only window that can be too much at a certain rpm or boost level and then not enough at another.

Parts needed:
bottle of some kind
small inection orfice (I used carb accel pump discharge nozzle)
Ebay cheapo boost controller
a few feet of windshield washer system hose or vac hose
fishtank one way valve
vac line "T" to tap into manifold pressure hose to injection pump
zip ties

I used a carb accel pump dump jet out of a Solex one barrel carb of a VW bug. I drilled the orfice to "60" (.6mm) and that has worked out to be a good size. I could probably size it up another .05, but I dont want to go too big and then have to get another one tio replace it. For me, the 60 made a noticable difference and there is no clatter when its running with it on. A nice side benefit besides the intercooling is that it will help clean your combustion chamber as the water turns to steam. Also, it really helps clean up the smoke! I went from a haze of smoke at 15lbs boost to absolutely nothing under full boost at WOT with the injection setup. Here are some pictures:

http://i1004.photobucket.com/albums/...3/101_0009.jpg
http://i1004.photobucket.com/albums/...3/101_0008.jpg

I positioned the orfice just ahead of the compressor wheel so the turbo can help atomize the water. Also, you may notice inline there is a one way valve for a fish tank. That keeps it from bleeding water into the intake when the water level is above the orfice. It is installed in the direction of flow, but provides just enough resistance with the little rubber flapper inside to keep it from bleeding. I've also incorperated a inline bost controller to regulate the boost level that the system starts to flow. Without it, you would drain your tank in no time as these cars run a couple to 5 lbs of boost on the highway. I have the system tuned to start to flow at around 8lbs off boost, so regular highway cruising there is no flow, but as soon as you lay into the throttle it begins to work. As the boost increases, so the does the pressure in the bottle and therefore the flow does as well.

Here is a diagram:http://i1004.photobucket.com/albums/.../watermeth.jpg

I used an old windshield washer bootle. You can use anything as long as it can be pressurized.

Jason

great idea, I like it a lot.
one day I will try it.

Did you notice what kind of power increase?
That idea really interests me.
I have soon my IC assembled (only minor work left)
and been thinking too about water/alcohol injection to my car,
since at this moment i dont have money required to buy bigger turbo and
"liquid head" to pump. Tho with IC there comes problem,
since it will chill water steam back to liquid form of water,
and thereafter just fall in bottom of IC.

Nice job anyways, absolutely!
I will re-desing your idea a bit, to work on IC, and also some minor modifications.
But, keep going, smoke your neighborhoughs!

I noticed a decent increase. Also, using windshield washer fluid has water and alcohol, so the alcohol is being burned as is just like adding more fuel. I don't know if you would really need it if you already have an intercooler, I'm still all hot air! If you were pushing lots of boost, I think it would help out as the intercooler can only do so much, but at 10 to 15lbs of boost I cant say for sure if it would help on top of the intercooler thats already there. The orfice I have is pretty small and it sprays the water in directly at the turbine wheel. That chops it up even more, I don't think it would collect at the bottom of your intercooler but who knows. It should be in water vapor form untill it gets into the engine, it turns to steam during combustion.

Jason

I think i will add nozzle straight to intake manifold, a bit towards
"sucking ditches".
So opening valves (see jbg i learned new words) which cause continuos
suction in hand with about 1bar turbo pressure,
should keep injected water on move, so it wont form liquid on bottom of the intake manifold.

Damn.. Some problems if i think to locate nozzle straight to intake manifold.
Pressure there is too 1bar, same pressure which tryes to push liquid from water container to manifold.. So that wont work, atleast wont work so at it would be any help. Need to think different way.

Ive been thinking this thing since i saw that topic today.
Ive made my own mind, and in my opinion,
that system pretty much loses its effective with IC,
and its harder to pull liquid throught nozzle to intake system like Jason did,
since it condense in bottom of IC due to high lowering on temperature in intake air.
Even it wouldnt fall on bottom of IC, i believe it would form beads in tight ditches of IC.
But if we would, with same Jason's system place nozzle straight to intake manifold, the fact at theres same pressure on intake manifold than the one which tryes
to push the water/alcohol mixture throught nozzle, that in my opinion dont work at all, or just forms water beads in head of nozzle.
And with normal fuel pump you cant push so much fuel to engine at it overheats due to that and brokes, so its very little power reserve to get.

But, as everyone knows from his/her own experiece, d24 powered engine is really slow and dead until turbo starts to spool. Now to my idea.
Its not totally ready yet, and might be so at some of you will
solve few problems i have ahead before me
Yes, we want to feed that water+alcohol mixture before turbo starts to spool to compensate it. Remember this on that point: when we have overfeed on engine, turbo will start to spool on earlier rpm's due to excessive exhaust gas.
When we inject that mixture to engine, it will allow more percent of fuel to be burned, like turbo pressure does, therefore adding more power.

But, we cant use that "under pressure" richer, so we have to raise the overall fuel feed.

Back to thing. Thats moded from Jasons original system.
We make pressurised tank, pretty much like Jason's old windsheild washer bottle, but that one needs to last continuos pressure, im thinking of something between 2-5bars. To form the pressure on tank, we will add one-way ventil to hose which comes from turbo to tank, so the pressure in tank wont be depent of turbo pressure. And pop-off ventil to tank aswell, to open when it reaches prescribed pressure, like 2bar, so the tank dont blow up.
Now, pretty much like Jason does, we will lead hose from tank, but this time, to intake manifold, where we have drilled hole, and assebled adapter to nozzle. But, somewhere to that hose between tank and nozzle, we will add magnetic ventil, which will be closed when electricity circuit is open, and opens when electricity circuit closes. Sensor will be placed in back of the gas pedal, on certain point, just need to build rack to it. So when we pull full throttle, or where you put your sensor, electiricty circuit closes, opens magnetic ventil and allows water/alcohol mixture to travel to nozzle and spray out from there.

Problems i have at this moment

How we will open the electricity circuit when turbo starts to spool and make pressure? Most likely we would want solution which would allow us to adjust at in what pressure electricity circuit opens. Ive been thinking at we could use one kind of system, where is a spring and some kind of plate on cylinder, where comes the same pressure what turbo makes. So when theres more pressure, the plate, where pressure comes, would push the spring "down".
And somehow that way cut electricity circuit. Maybe by pulling electricity circuit to magnetic ventil through cylinder, but when plate is pushed down, it would lead electricity circuit to "land" and that way cutting it before magnetic ventil. OF course we could at manual switch where to open circuit, at
magnetic ventil shuts down, but it would be manual, and noone of us dont want to play with some switch at all, so lets make it automated.

Im now really thinknig about this system over night, most thanks to Jason for original idea. Feel free to comment and please, give additional Ideas how to carry out so at electricity circuit opens when pressures come, to shut down the magnetic ventil and closing water/alcohol mixture flow to nozzle.

WOW that was alot to take in. What your describing would really be more fitting for a Nitrous system, like a small 25 shot to help the turbo spool. The water is to help cool the intake charge, as well as add some small amount of extra fuel via the alcohol. However off boost I don't think there would be any advantage to injecting any water into the engine.

Thats not so say it wont work, but sounds like alot to do!!!


Jason

Well, as you know, water is H20.
So when it "burns" it will give up oxygen.
There we face the fact at it works just like turbo: by adding more
pressure, we will get more oxygen at same capacity, therefore allowing
more fuel to be burned. Vaporizing water does the same,
we get more oxygen to same capacity, but without turbo.
And continuos suction in intake manifold will keep that water/alcohol mix
to continuosly flow to cylinders, so we might need bigger nozzle than you used,
to keep enough mixture going to all six cylinders.
Maybe we could even use multiple nozzles, drilled in side of intake manifold.
That was good addition too Jason, at it will cool intake manifold,
and cooler intake manifold we have, the lower is temperature in air,
and therefore again, we get more oxygen per cylinder to same capacity.
With metanol theres one problem: it will explode way before diesel will,
causing knocking to pistons, since pistons are still coming up.
So we cant use too strong mixture.

Of course we could also insert on/off switch to electricity circuit to
give us change to spread mixture via nozzles everytime we want,
so we make it half automatized.

I thinked about the automatic switch on night, but well, no
usable idea yet, need to keep thinking.

I'm using windshield washer fluid so the alcohol content isn't very high... To do direct port like you are talking, you would need a pump to run it. At that point it would probably be better to buy a kit like coolingmist or develsown as they come with a high pressure pump and they also sell nice digital proportional controllers to ramp up the injection quantity however you tune it.

Jason

http://i576.photobucket.com/albums/s...vesiruisku.jpg

My idea as picture to clear it up.

Idea there is same than in your Oil separator- design something unique,
and save hundreds of dollars/euros whatever.

I see your idea above, how are you going to get the system to flow by boost pressure? The pressure in the manifold will be equal so there will be no pressure difference to force the water in.

Jason

Yes there. Thats why we inject it before turbo starts to make pressure.
Until turbo is on full work, we have pressurised tank, whoes pressure
will overpass pressure on manifold.
Actually.. That way we could do "stage1" so at it will flow when turbo pressure
isnt up, and add mechanical switch, which controls magnetic ventil on pipe line, which will be controlled via
gas pedal, so at everytime you smash it to floor, it will open
pipeline, so it flows. When turbopressure is same as in mixture tank,
flow will stop. And we can add timer to magnetic ventil, so it will be open always like 6secs (just idea to clear my thing) and then it requires new signal to open again for next 6sec, and signal will be given when you take your foot out of gas pedal, and again smash it to floor. Simple.

Damn i just solved my problem how to control that magnetic ventil.
Solved it on previous post. To roll, where the wire from gas pedal comes, will be assembled copperbolt, which will be electrified, we will give really small controlpower to it, under 3A.
Now we imagine the roll is on "sleep", so gas pedal aint pushed, and the wire isnt turning the roll yet. Some way ahead of the bolt, is counterpart,
which will counter to that bolt. When the bolt slips over it, it will close electricity circle, and that small controlpower will wake kemo-switch (kemo-switch is like turbotimer, it will give power from battery for selected time, but with it we can use like anything to give it signal, like that 3A control power, which will "flow" only way under 1sec, but it is enough, it will activate the kemo) which will control magnetic ventil. Damn, solve was ahead my eyes and i didnt make it until now. That thing can only work (selfish).

Please point me where im going wrong, or what wont work,
i cant notice all myself, for me that looks "perfect", stage1 plan.

I don't know how much power or help it will be off boost but its worth a try I guess. Seems like alot of work for an unproven system though. It would be nice if you could set up a pressurized tank first with some simple switch, and pressurize it with an air compressor or something before you go to all the work to setup/wire/fab the system only to find it doesn't make the difference you hoped for.

Jason

Since you are already using a windshield washer tank, what about using using one that has a built-in pump? Would that supply enough pressure?

It might, but I like using boost pressure so the flow is proportional to boost. More boost equals more flow, where the electric pump will only pump a given amount which may be too much or not enough depending on boost and rpm.

Jason

Except using turbo boost to pressurize the tank means there won't be any flow since the pressure in the tank will be equal to or near the intake. Then in best case the low pressure won't atomize worth anything.

Thats why I was injecting it before the turbo. No pressure there. I got a more noticeable increase in power from the front mount intercooler though. I think I would do that first. Back when I did this, it was a cheap setup to try out and see what would happen.

Jason

I think that spraying this fine mist of water/methanol will destroy your compressor wheel. It seems that water droplets contacting blades spinning at near 100,000 RPM would etch them over time. Have you had any problems with the compressor wheel or is this the motor that you are rebuilding?

No problem with the compressor wheel. Injecting water/meth pre turbo is common practice, especially on draw-thru carburated turbo setups. Obviously the best results would be direct port, and I'm sure over time the water will errode the compressor wheel, but in a draw through setup for example, so will the fuel hitting it. I have never seen that wear out or ruin a compressor wheel on a turbo. Dirt is a much bigger problem. That being said, I'm sure someone somewhere has ruined a turbo from injecting too much or doing it for a long period of time, but it would have to be a hell of alot of miles of use, and we are only talking about injecting this for a few seconds at a time at the most.

Jason