Water injection... Where and why...

[Henny 0]

Just found this interesting article (and a load of others I'll post up once I find if they're relevant or not...)

 

From Autospeed

Water Injection - It lets you safely develop more power with nearly no running costs!

by Julian Edgar

 

 

Water injection is a technology that is nearly as old as the car itself. However, like many automotive technologies, it has waxed and waned as fashion has dictated. Water injection has the ability to suppress detonation, allowing the use of higher cylinder pressures. It is easy to control and relatively simple to install. In times of tight emission controls, decreasing fuel octane and rising petrol costs, water injection is one of the best ways of controlling detonation. And it has another major advantage over taking other approaches - the 'fuel' is available at almost zero cost!

 

 

How it Works

Water injection is used to suppress detonation. Detonation occurs when the flame front does not burn progressively across the combustion chamber but instead explodes into action. This causes a massive and sharp increase in combustion pressures which can damage pistons, rings and even heads. Detonation can sometimes be heard as a 'tink, tink' sound coming from the engine. The piston and head shown here has suffered severely from detonation.

 

Water injection works in three ways. Firstly, when the water is injected into the intake system prior to the cylinder head, the small droplets absorb heat from the intake air. Water has a very high specific heat rating (it can absorb lots of energy while only slowly increasing in temperature) and so the intake air is initially cooled. Next, the small drops of water start to evaporate. Water has a very high latent heat of evaporation (its change of state absorbs a lot of heat) and so the intake air charge is cooled still further. Finally, when the remaining water droplets and water vapour reach the combustion chamber, steam is produced. This acts as an anti-detonant and also keeps the interior of the engine very clean, so preventing the build-up of carbon "hot spots".

 

Water injection was first experimented with in the 1930s. At the time it was discovered that detonation could initially be prevented by enriching the air/fuel ratio. As cylinder pressures rose still further and that approach ceased being effective, the injection of water into the intake air stream was found to prevent detonation. Interestingly, the detonation remained suppressed, even if the air/fuel ratio was then leaned-out. This occurred because the excess fuel was being used to cool the combustion process. When water replaced fuel in performing this function, less fuel was then required.

 

This has major implications for both emissions and fuel economy at high engine loads. In fact Saab on some of their recent turbocharged cars has used water injection at high loads in conjunction with leaner air/fuel ratios to reduce emissions output and improve fuel consumption. To put this another way, at high engine loads it is possible to reduce the amount of fuel being used, replacing it with water without sustaining any loss of power!

 

 

Always Water?

While I have referred to 'water' injection, many systems add a 50/50 mix of water and methanol, or water and methylated spirits. Research carried out during World War II indicated that pure water is best at suppressing detonation, while a 50/50 mixture of water and methanol permits the greatest power output before detonation occurs. One reason for this may be that the alcohol burns more slowly than petrol, so causing peak cylinder pressures to occur at a later crankshaft rotation, increasing torque.

 

The question of whether a water injection system can increase engine power is a contentious one. While the intake air will be lower in temperature (and so denser) when a water injection system is operating, the presence of an increased amount of water vapour in the air means that there is less room for oxygen. It is for this reason that dry air (that is, air with a low relative humidity) can allow an engine to develop more power. So when the air is cooler but its water vapour content is higher, will more power be developed? If no changes are made to air/fuel mixtures, theoretically the two factors almost exactly cancel each other out.

 

This means that if water injection is used without any changes made to the tuning of the engine, improvements in power are possible but not probable. However, if the engine air/fuel ratio is leaned out, or boost is increased, or the ignition timing is advanced, more power is very likely. Supercharged aircraft engines using water injection had mechanisms that leaned out the air/fuel ratio simultaneously with the operation of the water injection. However it is very important to note that making random changes to the air/fuel ratio and ignition timing at high engine loads can be very dangerous for the health of the engine. Such changes should be made with care - it is very easy to blow up a forced induction engine with random leaning of the mixtures and/or ignition timing changes!

 

Both methanol and methylated spirits mix well with water when it is required that a mix be injected. However it is important to note that both of these mixtures are inflammable and so the anti-detonant injection system's storage container, pump and lines should all be designed and installed with the carriage of an inflammable liquid in mind.

 

Note that it has been suggested in some circles that the water can be directly added to the petrol by using a solvent such as acetone. However, I have not heard of anyone actually doing this!

 

Water Injection Systems

 

A water injection system should:

 

Distribute the water equally to each cylinder;

automatically start the water flow prior to it being required;

have positive shut-off (eg via a solenoid valve) when water injection is not required;

either warn the driver or decrease engine power (e.g.- by dropping boost) should the water supply be exhausted;

be very reliable.

Many aftermarket water injection systems do not satisfy any (let alone all!) of these criteria.

 

To be most effective, a water injection system should add water in proportion to the changing airflow. In other words, the flow of water should match the flow of air, with small amounts of water being added at low loads and high amounts at high loads. If very accurate control of the water injection quantity is available, maximum water flow per cfm of induction air should occur at peak torque when cylinder pressures are at their highest.

 

The water should be injected in as fine a spray as possible. Doing this results in each drop being smaller, increasing the surface area to volume ratio and so promoting evaporation. The smaller drops are also less inclined to fall out of the air, wetting the intake manifold walls and perhaps then being distributed unevenly from cylinder to cylinder. A small droplet size means that a high-pressure pump and a well-designed spray nozzle are required.

 

UK company URL at http://www.aquamist.co.uk produces some very sophisticated water injection systems - probably the world's best. The company has developed their own pumps which work at high pressures and low flows. The pumps use an approach a little like a bicycle pump. Water is drawn in during the induction stroke of the solenoid-like pump, then pushed out past a valve by internal spring pressure. The stainless steel armature pulses in this way 50 times a second, delivering up to 160cc a minute at over 70 psi. The pump has built-in electronics to control this pulsing, with a 0-12 volt input control signal able to vary the flow. While URL use a sophisticated ECU to control some versions of the system, the availability of the control signal input means that the output of the airflow meter or MAP sensor could probably be adapted for the same purpose.

 

An alternative to a pump is to use boost pressure to force the water through a nozzle. If this system is adopted, the spray can be used only in a forced induction car with the water introduced prior to the compressor. A very special nozzle is also needed if the spray is to be sufficiently fine to pass through the compressor without long-term damage occurring. People using coarse droplet water injection in front of turbos have reported that over a period of time the edge of the compressor blades develop a serrated edge - presumably from the impact of the water droplets.

 

The injection of water can occur at a number of different points within the intake system. In a naturally aspirated car, the nozzle is usually situated prior to the throttle body. In a forced induction car, the nozzle can be situated:

 

before the compressor,

after the compressor

after the compressor but before an intercooler

after an intercooler.

URL suggest a nozzle position just prior to the throttle body for road cars, while the supercharged aircraft of many years ago used up to 18 nozzles positioned around the supercharger exit diffuser. Testing of the two systems discussed below indicated that the best nozzle location should be found through experimentation. The amount of water that needs to be added to an engine is also best assessed through trial and error. If the flow of water is initially high and then is slowly reduced, this approach can be done quite safely. However, testing on aircraft engines indicates that the mass of water required to suppress detonation is 20-30 per cent of the weight of the total liquid charge (that is, the water plus the petrol) being consumed. The system should be configured so that water is only ever injected when there are high intake airflows.

[Kevin Bacon 5]

Finally, when the remaining water droplets and water vapour reach the combustion chamber, steam is produced. This acts as an anti-detonant and also keeps the interior of the engine very clean, so preventing the build-up of carbon "hot spots".

 

Very true. I used Aquamist on my 16V Turbo for nearly 3 years and when I lifted the head to fit a decent metal gasket (the standard one blew at 140mph!), the piston crowns and combustion chambers, valves etc were all immacualte with no carbon at all.

[Cougar 0]

I wouldn't do it on a G60 though unless you have the boost return removed. I can't believe that water would be kind to a Glader.

[keithmac 0]

I run Aquamist instead of an IC on my car, it has the boost return deleted, as you say the G60`s are fragile enough without water!

[NickVR6 0]

Humour me p,ease on an idiotic question guys. What would be the result of fitting and running this with the 50/50 mix on a normally aspirated motor like the VR?

 

I presume lowering the intake temps is allways good news, or do you need the forced induction to mix things up/excite the air enough for it to work. I suspect the answer will be that it will rpobably work but have little or no benefit at all on NA.

 

No laughing please! ??

[gibber 0]

~It does actualy work if you are running high compression.

[billinjah 0]

I run Aquamist instead of an IC on my car

 

 

i think i might try this first before i shell out on a new ic.

[samg60 0]

Noticed quite a difference on my car when running 70% water and 30% methylated spirts, I have it coming on at about 15psi and when it does kick in there seems to be much more torque at about 5000rpm it can be quite scary as you suddenly get more torque steer

[billinjah 0]

meths eh? does that mean there is another use for besides drinking it?

[Henny 0]

it's also good for lighting BBQs especially when you add in some Nitromethane... ;) :twisted: :onfire:

[markmcie 0]

Noticed quite a difference on my car when running 70% water and 30% methylated spirts, I have it coming on at about 15psi and when it does kick in there seems to be much more torque at about 5000rpm it can be quite scary as you suddenly get more torque steer

 

That sounds pretty awesome. Were is the water injected into the system ? Do you tap a hole into the TB ?

[Henny 0]

markmcie, and there's the big question... There's some debate (mainly started by me :oops: ) as to where the best point on a G60 is to inject the water... Just about everyone agrees AFTER the intercooler to stop water condensing in there, but after there it gets a little confused.

 

I thought (although I don't know that's part of the point of this thread, to gather info) that the best place would be just before the CO pot so that the ECU is aware of the sudden drop in inlet temps and can adjust the fuelling accordingly preventing lean running. Other people have said to insert it just before the throttle body with others injecting directly into the inlet manifold to avoid water vapour being sent back down the boost return and into the G-Lader... :?

 

I'm still reading up on a load of info I've dug out, but the general concensus so far is "trial and error" to find the best place... :|

[olly elworthy 0]

markmcie,

but the general concensus so far is "trial and error" to find the best place... :|

 

trial and error and the G lader? :roll: not a good mix, however this is new ground,, :thumbleft: for starting this thread and getting some reasearch and ideas thrown about,,,

To be safe i think a boost return delete is unfortunatley mandatory,,mind you that wont be so bad if we can get rid of the squeeling noise on part throttle, the carbon cannister take of lugging looks as if it may ovecome this one with a nice length of pipe on the bypass outlet,,,

may be worth going on the pitstop developments forum as steve and co have fitted aquamist systems onto a few G40`s,,, :wink:

[Henny 0]

yeah, don't like the thought of blowing up one of my G-laders... I've blown up just about every other part of an engine in the past and don't want to blow up something that's getting as rare as the G60! :oops: :lol:

 

I hadn't thought about Pitstop, I'll go and have a nosey... ;)

[samg60 0]

I am currently running my injector about 6 inches before the co pot as this was the only,

feasable place,

When My silencing kit arrives I will be relocating the injector to just before the throttlebody.

although my kit is not from the same company as the recent group buy

mine was (http://www.coolingmist.com) it is essentially the same kit and, the main problem I have with it is the angle of the

injector, and I will soon be purchasing and angled injector housing, so that I can angle the direction of flow with the

air flow

[Henny 0]

Interesting, I've just got some interesting info off Coolingmist's (and also Aquamists) website before I saw your post... 8)

 

Quote from aquamist.com

Where do I place the water jet?

Normally immediately after the intercooler unless the intercooler suffers from heat soaking such as the type that is fitted on top of the engine (Subaru, GTI-R, Toyota Celica and etc).

 

Surely if the jet is placed before the intercooler it will have better cooling effects?

Not quite true. If the air entering the intercooler is pre-cooled, the cooling efficiency of the intercooler will drop due to the smaller temperature differentials between the ambient air and induction air within the intercooler core. Secondly, there is also a possibility that the hot air from the turbo may cause unnecessary vaporisation of the injected water thus taking up precious volume that was intended for the charge air.

 

Should the water jet be facing the direction of the in-coming air?

Two possible locations, if the jet is to be installed along an induction pipe, point it at 90 degrees to the direction of the air flow. Position the jet furthest away form the throttle body as this will enable better mixing and intercooling properties. If the jet is installed at the end-tank of the intercooler, point the jet in such a way that it has the greatest cavity to accommodate the 90 degree spray pattern.

 

Quote from coolingmist.com

Standard Injector Mounting

 

Depending on your vehicle the installation can be different. The first thing you need to do is decide where to mount them. For a dual stage we recommend that the larger injector be mounted furthest from the engine and the smaller one be mounted close to the engine.

 

If you have an intercooler you can mount your injector prior to the intercooler or just after.

We recommend to install the injector after the MAF sensor of your car (if you have one). If you have to mount before make sure you have it far away so it can evaporate completely.

You can drill a hole to install the injector or tap the injector.

We recommend that you install the injector AFTER the turbo or supercharger unless you have no other choice.

[Kevin Bacon 5]

I had mine mounted just after the intercooler on the my 16V T like they suggested. Worked well.... although there was no CO pot to consider :-)

 

Not sure where to put it on the VR. I've got an 1/8th brass fitting already attached to my charger outlet hose near the throttle, so might shove it there initially and see how that works. Should be OK with one of the finer nozzles.

 

The Aquamist people are pretty hot actually, always helpful on the phone and quick to sort problems.

[gibber 0]

Would I be right in saying that the Co pot is not realy a MAF tho?

 

On my VR I would never fit the injector before the maf.

 

We recommend to install the injector after the MAF sensor of your car (if you have one). If you have to mount before make sure you have it far away so it can evaporate completely.

[keithmac 0]

CO pot also houses the IAT sensor, it can stand being covered in water quite happily from my experiences! Wouldn`t think covering the MAF would be a good idea as you say.

 

My thinking is it has to be before the CO-pot (so the ECU sees the denser air) and has to have enough distance to cool the charge. My nozzles tapped into the RSR outlet, although the intake run is very short. Run it for 6months like this and it`s been fine.

 

If you use the Aquamist pump be careful what % alky you mix in, or buy a seal kit for the pump that lets you run high alky content.

 

I`m toying with the idea of buying a DDS3 readout/controller, monitors for faults and can dump/limit boost in the event of a failure.

 

N/a cars can benifit by being able to up compression or timing advance, you can also run leaner mixtures to get the best mixture for maximum torque, which is normally not possible due to knock.

 

My valver always used to respond better on cold damp mornings due to denser air.

[Shaggy 0]

Just to add to this thread but what do i do if i got the dual stage kit with the that has two injectors? i am thinking of putting one just before the Co pot ( larger Jet ), and maybe another one ( smaller jet ) maybe just after the intercooler, bearning in mind i already have a g60 intercooler setup now so intercooler heat soak isnt such a problem. would this work? i dont really want to start drilling holes if its not in the right position now do i... 8)

[Kevin Bacon 5]

On my VR I would never fit the injector before the maf.

 

Absolutely, goes without saying. If the water didn't upset the wire and air temp reference sensor, then the ultra dense charge would throw the MAF out of range instantly causin the ECU to panic and dump excessive fuel into the chambers.

[billinjah 0]

My thinking is it has to be before the CO-pot (so the ECU sees the denser air)

 

 

this would seem sensible!

 

this is getting more and more confusing! where the hell should i pt the damn thing!!

[gibber 0]

Wow me and Kev agreeing :lol:

 

mike_g60 I would say fit the larger jet further away from the TB.

 

Im itching to get my G60 cabrio back from the body shop now so I can play around fitting my second injector, hehe

 

I still stick to what I was sayin about fitting the jet after the tb in the g60 manifold, you wont get an even spray into all cylinders which will cause problems.

 

So I got one just before the Co pot, and I can say that it works but that is the only place I have tried so it mite not be the best.. Now I dont know where to put the second, just before or just after my huge intercooler.

 

As for removing the boost returm, wen you lift off the gas and the boost return opens, the solonoid on the kit will close so it will only be the small amount of water that mite get into your charger. I can see that being a problem.

[keithmac 0]

Post a pic up of your engine and we`ll make a decission for you :wink:

 

Have a good look round on the forum at http://www.aquamist.co.uk, a mine of info in there!

[Chris M 0]

This is what i've come up with (see pic) and its going just before the CO pot i'm not sure about them both being at the same point but i can always make another for somewhere else