Water Injection Theory

[Gollum]

So me and some buddies have been talking about water/meth injection systems, particularity the best ways to apply them to certain applications to best suit their needs and also make it easy to tune/work with and live with.

 

There's been much discussion on WHERE to spray in the intake system, and a friend with a STI is going to add water injection soon and we're both in agreement...

 

 

...before the turbo. There isn't a ton of information on the net, which is why I've come here. The consensus is that water injection system of old sprayed too heavy of droplets and could damage your compressor wheel, which is why it wasn't common practice. But modern nozzles with high (3 bar) water pressure have a fine enough mist that they're relatively "turbo safe" (we hope). But the theory goes that if you cool the charge before the turbo then your turbo efficiency goes up. Spray enough water and it's almost pointless to have an intercooler, or most certainly a large one. Some guys are "claiming" as much as a 34-40% turbo efficiency increase (also directly related to the max HP/Flow of the turbo).

 

I plan on going megasquirt eventually still (been saying that for how many years it seems?), and I'm thinking that I might be able to get away with running without an intercooler at elevated boost levels. I'm only going to be shooting for 300-350 wheel hp or so. I want to keep my tire size between 225-245 and not have way too much tire for power, but that's a wholenother topic.

 

Please, everyone, chime in. Let me know what you think about where to place water injection and why. And if you have experience with water injection please share details!

 

Side note~ I'm also interested in water injection for fuel economy reasons as well, but I still haven't quite figured out how that'd work for tuning it in MS.

Edited June 19, 2010 by Gollum

[Snailed]

I always put the nozzle(s) after the inercooler and have had really good results. Getting the water into the combustion chamber seems to make the most sense to me. Waters latent heat of vaporization is so much higher than fuel so using it to take heat out of the combustion chamber directly, rather than to just cool incoming charge air, makes sense to me.

 

IF your goal is to make the turbo more efficent I really don't have any hands on experience with spraying pre-turbo but I'm suspicious of claims of compressor efficency in the range of 100%.

[Zmanco]

Hmmm, interesting question. It seems to me that injecting water upstream of the turbo to improve the maximum flow of the turbo makes sense - but this would only matter if the turbo was undersized for your target power levels. For example, if you were trying to push a stock 280zx T3 turbo into the 300+ HP range, then I could see a benefit to injecting upstream of the turbo.

 

If the turbo already has enough capacity, ie. for a given pressure ratio (PR) and flow, it's still within a reasonable level of efficiency as shown on the map, then I would think that practical considerations would favor spraying after the turbo. For example, even with higher water pump pressures, all else being equal, spraying into a warmer air flow would probably result in more complete vaporization and subsequent reductions in temperature.

 

So I think you have to take into account how well the turbo fits the application before you can determine which approach is best.

[spork]

I hope I don't come of sounding to rough, but have you ever seen what dust does to a turbo impeller? I'm not talking about sand or large particles, but very fine dust. I see this from time to time on the semi's I work on. Just a slightly loose clamp after the filter and the finest dust you've ever seen makes it in and eventually destroys the turbo. I really imagine that the droplet size is going to still be too big. I run meth injection on my car, and use a system that runs around 120-200psi depending on how you have it set up (Devils Own). I spray mine in the runners, but I am spraying very large amounts because I'm also using it as fuel since my 52lb injectors are too small to keep up. Good luck with your experiment tho, I'm interested in hearing about the results...and I hope I'm wrong, but I'd like to see before and after pictures of the turbo over a couple 5-10 thousand miles of use.

Edited June 19, 2010 by spork

[Gollum]

For those wanting to kill some time, here's a good thread on pre-turbo results when it comes to temps

 

http://www.aquamist.co.uk/vbulletin/showthread.php?t=531

 

Spork, no worries, you sound as rough as a baby's butt. One factor that I think helps this work is that the vapor we'll be spraying is already practically atomized already. The dropet size is already so fine it's practically vapor. That being said most of the guys doing it are making sure to be spraying at the center of the impeller, to help prevent possible damage.

 

I think another factor preventing turbo damage is the nature of water. Dust and debris are small pieces of things like ROCK. Even on a microscopic level it's a HARD objection. If someone throws a pebble at you at 50MPH then it's going hurt a bit. When spray hits your face while ridding a seadoo it only stings a little.

 

There's obviously surface tension on water, and people say that "hitting water at 80mph is like hitting concrete" but I think mythbusters did a decent job of killing that farce. The bottom line is that water IS soft at any speed when you compare it to other objects. Now, obviously water CAN be hard, like in water cutting tools. But these are firing a SOLID stream, not a stream the size of droplets. They're also firing at insanely high speeds. Not exactly apples to apples on that front.

 

 

 

Oh, and something I read today was a guy who uses pre-turbo injection on his EVO and mentioned that he gets much faster spool time. Theory being that since normally the turbo exit temps are as high as 300F that spraying into the turbo lowers the temp so much, that the increase of density of the ouput means you get into boost that much sooner.

 

 

And I think I agree to an extent that this makes most sense in cases that your turbo is undersized. But that being said, does anyone have turbo exit temp numbers for even the most oversized turbo? And how many people are brave enough to run without an intercooler praying that their "turbo isn't working too hard"?

 

My friend that's going to install WI on his STI is already nearing the limits of his stock intercooler. He's at around 350 wheel HP. He's already found the limits of the stock turbo and has replaced it. He found the limits of the stock injectors his first day tuning, and those have been replaced. He plans on replacing the intercooler with another top mount, to keep it factory looking (we are in CA after all) and though it will remove some of the flow restriction currently being seen, it will still be a small intercooler unable to extract extreme amounts of heat.

 

In my case I want to add it in order to possibly avoid the need for an intercooler AND using a small turbo (stock T3). That T3 has proven to get people people to 300 wheel HP, but you need a larger intercooler and a rich mixture. I'm not against upgrading the turbo, but I don't plan on going over a small T3/T4 when I do. I figure that trying to spray after the turbo will be fighting a loosing battle. The turbo will eventually heat up more and more and the water will have a harder and harder time keeping up. If I spray preturbo the effort will be to keep the temps of the air AND the turbo manageable to being with, even under heavy track conditions.

 

Well, that's all I have time to write right now. Keep the input coming everyone!

[johnc]

Here's some of the original research papers on water injection:

 

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930093206_1993093206.pdf

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930093247_1993093247.pdf

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930093243_1993093243.pdf

 

In all current piston aircraft applications water injection is done after the turbo (if water injection is used at all). Most race teams run enough fuel that they don't need water injection at all and they generally avoid it as yet another failure point (Koji forgot to fill the water bottle again). The general trend is to properly design the engine and systems to not require water injection to control detonation.

[Bernardd]

The rx7club forum has a couple of good threads in the aux injection forum for preturbo water injection setups with diagrams and results. I would like to see a link to the 34-40% increase in efficiency with preturbo water injection. I would switch to that type of setup if it would save having to upgrade my h-trim to a gt35r. I think a combination of the two might be the way to go. I cannot find a link but I seem to recall someone explaining why air saturated with methanol will still have room to carry water.

[Snailed]

I have done a few supercharged vipers with water pre-charger..one of them made over 1000whp/ftlbs on pump gas and water. All the STIs I have setup ran one nozzle right before the throttle body and on stock turbo/topmount cars we got about 50 whp with a 50 water 50 meth mixture. I also had good luck on big turbo STIs and made high 500s on pump and water. Everyone has different experiences and I know I will just stick with what has worked for me. I still think taking heat directly out of the combustion chamber makes far more sense than just cooling the charge air. I'm more concerned with system efficency than just turbo efficency.

 

If you do go pre-turbo I really doubt there will be any damage to the compressor wheel, as long as you are running high pressures and using a modern agracultural style nozzle. I have never run a filter on my street turbo setups and after 50k miles on the stock turbo the compressor wheel was barely showing any wear on the leading edge. Check out the compressor wheels on some of the fastest cars in the world...they are often beat to hell and working just fine.

 

john..that's a good practice for sure, but for street cars, running race gas is just too expensive and inconveinet. Added complexity can change reliability of a setup but there are steps that can be taken to make it a reasonable risk. Many people try to argue that on the carburator vs fuel injection front and I always ask them which system they drive to work or give the wife to drive it's all about refinement.

[Gollum]

I too can understand why race teams wouldn't use it. I also see no reason for a race car that spends most of it's time in the power band to have a vtec type technology. There's lots of things the a street car might be more suited to than a race car. And I'm definitely one to believe in the "design the system the best you can to begin with" philosophy, but sometimes "band-aids" don't turn out so bad.

 

I should definitely make it clear though, we plan on running near 100% water on the STI. If we add anything it would be for corrosion resistance.

 

I'm also skeptical for HUGE HP limit gains on the turbo, but I could see 10% being feasible. I currently don't have links, I'll see if who I heard it from has any idea where I could find this "information". I personally called BS when I heard it, but at the same time we're going water injection regardless what's true or not.

 

So here's another question for you though. What does everyone think about a small nozzle that comes on around 3-4 psi and another small nozzle right before the TB to lower chamber temps? Might be worth experimenting with for someone like me trying to dodge the use of an intercooler.

[johnc]

What about using a fuel injector and associated plumbing/control if you're going to spray before the turbo? That should give a fully atomized spray. Also, a lot of the local autocrossers (especially the turbo guys) are switching to E85 and getting rid of meth/water injection.

[Gollum]

I'm 100% in agreement with the switch to ethanol. Some of the compression ratios guys are running on turbocharged engines is nuts! Ethanol is still a little hard to get a hold of around here though. Only station close is walnut creek, 15-20 minutes from most people in the bay area. If you don't mind stocking up and having spare barrels around I guess it's not so bad. That's kinda the upside to a separate tank for an alternate fuel. If you have a low level sensor you can just keep out of boost until you have time to refuel your boost gas (or whatever you want to call it).

 

I had originally thought that using an injector would work fine for water injection, but apparently people shy away from it because of rust issues casing injectors to seize, combined with the fact that apparently the high PSI nozzle systems work better for atomization. I'd like to see a direct comparison though.

[ktm]

John, I am running FJO Racing's setup which pressurizes the system to 120 psi and utilizes high speed solenoid valves that are PWM controlled. I have two 700 cc/min injectors (1400 cc/min total). I was spraying 100% methanol but am now running a 60/40 meth/water mixture.

 

Taking out the heat of combustion does not increase power, it allows you to run more boost or increase timing for a given boost level that increases power. Cooling the air charge increases power by the increase in air density. Methanol is more effective at cooling the charge than water and has the added benefit of increasing your fuel's octane. I am not going to get into which one is better because I don't know and it is the topic of MANY debates. The Subie crowd swears by Aquamist and their insistance on water injection. The Regal boys over at Turbobuicks.com swear by Alkycontrol and 100% meth. I was replacing up to 25% of my primary fuel with methanol when using 100% methanol.

 

I would love to run E85 in my car, but good luck finding local gas stations that carry it. It is has hard to find an E85 station as it is finding a station that carries 100 octane at the pump or a methanol supplier!

 

I would not rely on methanol and or water injection alone in-lieu of an IC unless you had quite a few redundant safe guards built into your setup. The IAT post-IC on my car is around 100 degrees F; after injecting methanol my IATs were down around 40 to 50 degrees (the pipe was nice and cold).

[Tony D]

Do the mass-flow calculations, and you will see preturbo is 'no free ride'---the mass of the air, while increased....also contains water. You can preinject before the turbo to the limit of 100% R/H on the outlet of the turbo. It must then go DIRECTLY into the engine as if any cooling of the charge air occurs condensation and liquid water will form.

 

The density increase from latent heat offsets the loss of air due to water displacing it, so you get more power.

 

If you had something that burned (methanol) then the latent heat issue is AIDED by the 'carried along' anti-detonant as it is burning as it goes.

 

In either case, your cylinders stay really clean.

 

Spearco Water Injection (2 Stage) on the car since 1979...

You know they made them for N/A's back then as well?

[Gollum]

It must then go DIRECTLY into the engine as if any cooling of the charge air occurs condensation and liquid water will form.

 

 

Yea, I was just reading an article today on RB Racing's website about how you can't water inject pre-intercooler. Something to ponder for sure, especially since there seems to be quite a few people injecting water pre-turbo that I'm pretty sure have intercoolers.

[ktm]

It really depends on the efficiency of the IC, the intake air temperature pre-compressor, the amount of water injected, the humidity of the air pre-turbo, boost pressure, the temperature at which the air charge leaving the compressor, etc. Look up psychrometric curves for a better understanding. It is not very straight-forward.

 

Basically, hot air holds more water than cold air. Air is hotter leaving the compressor and can hold more water for a given pressure. As the air charge drops pressure and drops in temperature, condensate forms.

[Tony D]

General Rule of Thumb is air at 120F holds twice as much air that is 85F. We get this in industrial compressors all the time. Design is to have air entering the second stage at 120 so no condensate removal trap is required. Some genius Facilities Engineer decides he's going to out-think the OEM and 'increase intercooling efficiency' so he taps off his chilled water system to decrease interstage temperatures.

 

Forgets condensation will occur. Formerly air passing into the 2nd stage was 85-100% R/H at 120F, it's now 100% R/H at 60 degrees. Ooops, 'water got out'...

 

And then after a couple of hours running, they get a vibration shutdown. Or worse. Then the phone call comes about 'your pos compressor'...

 

The 'eureka moment' is golden: Did it ever occur to you why we didn't provide a condensate trap on the 2nd Stage, Sir? At 120F we don't get any condensate dropout. You used electrically generated chilled water to 'increase efficiency' and while you increased air density you also increased electrical load on the compressor motor...so what did you REALLY get for that 'increased airflow out of the machine'? We can get you more airflow, and yes, we will want a larger motor. NOTHING is free! Well, maybe if you had 48 degree ultra-clean well water you could have eliminated the chilled water electrical cost... but you don't so let's talk about this rotor wreck you have precipitated by flooding the second stage with condensate, my esitmate is around $22,000 for the high speed rotor assembly, plus labor.... What was that you said about 'saving money' again?