Water/meth injection: a theoretical and practical discussion

[bobhatton]

Bob if you have the time to go into detail, I'd be interested to read about some of the setups you've built over the years and how these worked, because you keep dropping tidbits of info without ever really expanding on it. I do understand what you are talking about.

For instance, where would your injection points be if injecting separately? Pure methanol before the throttle plate and water directly into the manifold? That would seem to make sense based on what you've said. I don't suppose many people with road cars would like a tapped tank of pure methanol in their cars. So, for road cars, it is safer and more practical to inject together, so methanol can both cool the intake charge and increase octane, and water can provide additional cooling in the cylinder when evaporated there at higher temp?

The manifolds we use are "dry", they are not designed for a fluid to flow through them, so the distribution will not be very even.
To give the correct amount of water or methanol they need to have one injector per manifold runner.
Anything that you inject into the manifold will displace air, but what we are after is more air not less.

Water will absorb a very large amount of heat, it does so in the cylinders but in the manifold there is just not a large enough difference in temperature between the water and the air. It also has only fractions of a second in the manifold to do anything, just no enough time.
So with water it needs to be injected as close as it can be to the inlet valve.

With methanol it will absorb heat at a much lower temperature differential and much quicker, cooling the charge, this will increase the airflow in the manifold.
So Methanol should be injected right at the top of the manifold runners.
Now because methanol is a fuel the engine will be running a richer mixture, and even more so if it gets on to the IAT sensor as the ECU will also put more fuel in as it thinks it has more air.

With the JDM turbo engine cars we have most of the time they are running under the control of the knock sensor, you will not know that, it is all auto control by the ECU.
Put a water injection system on and you may well feel a boost in power, because there will no longer be knock and the ECU can go back to its design ignition timing and a/f ratio.

Put water into an engine without knock, because it will have reduced the temperature and therefor the combustion pressure the power will be reduced.

[JD]

Great, thanks Bob, that was a good read.

I (and many others with upgraded engine management) have my car mapped specifically for 98 RON. On my current budget, based on what you have said, my limited technical knowledge, and based on the numerous experiences of others that have done the same, or very similar, I believe a 1:1 water/meth mixture injected before the throttle body will be advantageous.

Since someone in my position is not developing and building a race engine, based on mountains of data, and developing any number of custom parts for such a build, I will just go ahead and see what I can achieve, mapping for the w/m when fitted. I will post the results up here so that anyone else that is interested in doing the same can reference and decide for themselves.

[GeorgeL]

Having jets on the inlet runner before the inlet valves is daft. The vacuum at idle is too large which means there is potential the water meth will be dragged in rather than injected at the rate required due to the large pressure difference.

[ashley]

I am still nervous about what it's doing to my IAT readings and whether I can believe them anymore if injecting pre-TB

[bobhatton]

Having jets on the inlet runner before the inlet valves is daft. The vacuum at idle is too large which means there is potential the water meth will be dragged in rather than injected at the rate required due to the large pressure difference.

A solenoid valve is used to stop that

[jimGTS]

I am still nervous about what it's doing to my IAT readings and whether I can believe them anymore if injecting pre-TB

if the sensor is getting coated, so are the surfaces of the IL and IC pipes, would this "coating" not help temps? ie air flow would be floating on these cold particles and not hitting the metal?

just thinking out loud.

[2mad]

I'm going to try and keep this thead alive because I still think 50/50 lowers intake temps , read ww2 wiki engines with water & water/meth injection wiki says boffins wanting to build faster fighter airplanes used water and 50/50 for higher boost and lower intake/fuel charge temps for take off and dog fights (6psi and 12psi with 50/50 spraying).


One of the wiki articles. http://en.wikipedia.org/wiki/Water_inje ... 8engine%29

The other wiki article .. http://en.wikipedia.org/wiki/War_emergency_power

Also this .. http://www.enginebasics.com/Advanced%20 ... ction.html

They used nitrous too, engine builders were proper hardcore in ww2

Kev.

[Martin F]

I am still nervous about what it's doing to my IAT readings and whether I can believe them anymore if injecting pre-TB

I just got my car mapped and my water injection is in the throttle body, the intake temps were almost identical, my ecu is retarding my ignition when temps go above 50 degrees. This is a safety measure on the ecu I was advised to leave in place. Timing was mapped more aggressively but the darn ecu kept taking over and taking away any good the water injection was doing. I was advised that if I sorted out the intercooling I would be looking at an extra 40 rwhp on this map, I could then go for for higher boost and even more power. I was also advised to get something to confirm my water injection was operating when accelerating. . So it's useful but if it fails with an aggressive map you could see engine damage.

[Martin F]

I was also losing a few bhp when it came on, but boost could then be raised more where it couldn't before, everything was looking peachy until the ecu iat sensor took over and gave the car the brick wall effect.... in hindsight I should have learnt from past mistakes and done a proper job of lowering intake temps the usual way, ie. Bootmount or large chargecooler. So water injection doesn't get my vote, a-for safety reasons and b-it didn't fool the mr2's iat sensor so in my case was off little use, apart from maybe a risky safety net...

[jimGTS]

I don't think it drops intake temps a bunch, I think it helps stop det/knock.

This is my experience anyhow.
I don't think you make anymore power, simply keep knock at bay.

Like for like, wi vs no wi, same power but with wi, less knock events.

If mapped should make more power due to potential additional timing.

[2mad]

I am still nervous about what it's doing to my IAT readings and whether I can believe them anymore if injecting pre-TB

I just got my car mapped and my water injection is in the throttle body, the intake temps were almost identical, my ecu is retarding my ignition when temps go above 50 degrees. This is a safety measure on the ecu I was advised to leave in place. Timing was mapped more aggressively but the darn ecu kept taking over and taking away any good the water injection was doing. I was advised that if I sorted out the intercooling I would be looking at an extra 40 rwhp on this map, I could then go for for higher boost and even more power. I was also advised to get something to confirm my water injection was operating when accelerating. . So it's useful but if it fails with an aggressive map you could see engine damage.

This why I have not installed 50/50 in my st205 (i would need to install in the throttle body, fmic on the way purely for water/meth injection) , you need to instal the jet 12 to 15" from the throttle body or you dont get any benefit, the iat does not know intake/ fuel charge temps are lower and the ecu acts accordingly, bad times (because it thinks you intaking 50c+).



Kev.

[2mad]

As far as engine damage is concerned, my car was tuned for 11.5 afr 50/50 off. So if w/i failed the iat would see a 50c spike and react also the knock sensor would also see conditions have changed and if they were severe go to bad gas mode
And a led light would illuminate on my dash.

Put alot of thought into this, just trying to enlighten the forum to my findings, best mod hp wise I have ever done and it only cost me £120 what a bargin

[2mad]

Some additional information, my car was running a hks vpc , so air intake temps were clamped at 25c by a resistor and fueling was calculated by a hks iat senor in the throttle body .
50/50 OFF the car ran like a mr2 if well modded , 50/50 on car ran like a champ even raced a Audi r8 in a straight line and wasn't embarrassed



We hit some roundabouts going straight over, i got my a*s handed to me (pardon my American lol ) they corner flat

[raptor95GTS]

if you're going to use water injection, you might as well do it properly and fit a progressive kit so you maintain the ratio of WI to fuel throughout the rev range of use. No point dumping the same amount of water / meth at 10psi as you do at 15psi boost, you'd either run too much at 10psi boost to get it right for the 15psi boost requirements.

Bob will argue where that kit should actually be fitted but I don't think he'd argue against a progressive kit. Then again he might, just for the crack of it

[2mad]

Mine came in at 0.4bar with full bost at 1.1bar and it pulled like a train from the moment w/i came in . If a progressive kit gives more torque and hp i'm in


Remember w/i is not a exact science when it come to what to inject, i've use lots of different calculators and got different answers . I inject 0.7mm jet with a aquamist race pump and it worked great.

[raptor95GTS]

er actually it is pretty much an exact science, the guys building the fighter engines did lots and lots of testing with different mixtures and ratios. One point they had water flowing out of the exhaust stacks of an aircraft engine and yet while it made not a lot of power, it didn't hydrolock either. Frank Walkers story is a good one Think they came up with 1/4 lb of water and a 1/4 lb of methanol to every 1 lb of fuel but don't quote me on it. Remember, these guys were not using the alcohol for fuel, they were using it to stop the water freezing at attitude . Fuel and boost they had more than plenty of


edited - I think it's 0.25 lb water plus 0.25 lb methanol per 1.0 lb of fuel

[2mad]

I used a 0.5mm jet and it worked fine I only switched to a 0.7mm jet because T.F.S and aquamists site said to . Changing jet sizes made no real word difference to afrs and hp . That is the only reason I say its not an exact science you just need to be in the right ball park with it.

You need, from testing to find what works best for your engine.

You need a calculator that takes into account, pump flow and injector size to calculate jet size . and all the sites i've used gave different results .

I'm shore the ww2 boffins would have this down to an exact science, but were on our own

[raptor95GTS]

indeed, for what we're likely to be using it for it's a big ball park we can play in, it's certainly something I want to add to my car. The fact that it cleans all the crap off the inside of the combustion chamber helps too

[2mad]

Agreed ^^ water/meth does so much more than coolder air charge temps (allegedly lol) , steam cleaning the head and combustion chamber is a big plus one

[raptor95GTS]

German engineers tried water injection (Wassereinspritzung) on their gasoline
engines, but with limited success. Germans, who were very good at building high precision pumps, had perfected direct fuel injection for their large aircraft engines. German engineers injected water directly into the cylinders as well. Since the water did not have time to evaporate and cool the induction air, the large cylinder inlet temperature reduction was not achieved. Frank learned of this while reviewing a report on a captured German aircraft engine.