Was looking into a water injection setup. I have read various posts and was wondering what people had to say about theres. My plans are to run the FP20G with the Ultimate racing 750 injectors. DO you think I can run that setup and get good results on the stock intercooler? Right now I'm at Turbo xs stage 2.5 and I have heard some people freeing up as much as 10-15 hp from switching over to WI.

Next, how is the install? I know you have to drill a hole in the stock intercooler but besides that what else is involved? Maybe someone has documented a install already? Also maybe someone can recommend a reputable kit that is easy to install and works well. Preferably one that comes with a tank already. :D

Thanks in advance.

pm'd ya :p


Jeff

Originally posted by NavyBlueSubaru
pm'd ya :p


Jeff

Jeff, you willing to shoot some info my way as well? I am curious as well.

Wouldn't mind a PM also jeff...:D

~Evan

Jeff from orlando why dont you just come out with your amazing info like you did in the bragging forum..

Lemme guess, this week youre pushing 400whp on 440cc injectors and you put koolaid mix in your water tank?

The best kits seem to be from Aquamist.

Don't go in expecting more HP without changing boost levels or tune. What WI will do for you is allow you to run a hotter tune, and more boost, safely, as it effectively raises the octane level of your fuel and reduces the intake air charge. Since you have an intercooler, you'll want to place the nozzle AFTER the intercooler, or else you'll risk getting puddles as the fins inside the intercooler will kill the atomization of the water mist.

The water actually takes up space that otherwise air and fuel would occupy, so in some cases people will lose a tiny amount of power, but for turbo cars it's a win - win. You can also mix in alcohol with the water which will act as a fuel. (I was informed Alcohol is not compatable with the seals in the Aquamist pump) but is used on other systems.

Matt

can someone point me to some sort of diagram/better explination of exactly what WI is? from what I've been reading it sounds kind of like you uses it in a similar fashion as NOS, but I must admit i know nothing about it, but want to learn. Thanks.

Go to the resources/faq on http://www.aquamist.co.uk/

Plenty of info there, also pics of a wrx with w.i.

~Evan

Originally posted by Porkchop-WRX
Was looking into a water injection setup. I have read various posts and was wondering what people had to say about theres. My plans are to run the FP20G with the Ultimate racing 750 injectors. DO you think I can run that setup and get good results on the stock intercooler? Right now I'm at Turbo xs stage 2.5 and I have heard some people freeing up as much as 10-15 hp from switching over to WI.

Next, how is the install? I know you have to drill a hole in the stock intercooler but besides that what else is involved? Maybe someone has documented a install already? Also maybe someone can recommend a reputable kit that is easy to install and works well. Preferably one that comes with a tank already. :D

Thanks in advance.

Here is some good install info from aussieinstlouis, I just went through this install a few weeks ago and he helped out alot.

Link (http://www.clubwrx.net/forums/showthread.php?s=&threadid=27256&highlight=aquamist)

I'm going to buy a water injection kit if I do good in Vegas in a couple weeks :devil:

Originally posted by I am mattjk
The best kits seem to be from Aquamist.

You can also mix in alcohol with the water which will act as a fuel.

Matt

DO NOT USE ISOPROPYL ALCOHOL W/ AQUAMIST The pump seals will be degraded by use of iso. Use Methanol at a mixture no more than 50/50 with water.

By the way, water injection is good. I get to run 87 octane gas while commuting to work and still have the TXS 4 knockless. Saves so much money in gas and my milage has increased 20-30%, depending on conditions. In a year with gas savings, I'll have paid for the Aquamist kit.

Originally posted by mlambert
Jeff from orlando why dont you just come out with your amazing info like you did in the bragging forum..

Lemme guess, this week youre pushing 400whp on 440cc injectors and you put koolaid mix in your water tank? :lol: :lol:

I run an Aquamist 1s kit and I love it. My install is featured on the Aquamist site here:

http://www.aquamist.co.uk/rescr/gallery/subaru-tm/tm.html

Originally posted by zaxrex


DO NOT USE ISOPROPYL ALCOHOL W/ AQUAMIST The pump seals will be degraded by use of iso. Use Methanol at a mixture no more than 50/50 with water.

By the way, water injection is good. I get to run 87 octane gas while commuting to work and still have the TXS 4 knockless. Saves so much money in gas and my milage has increased 20-30%, depending on conditions. In a year with gas savings, I'll have paid for the Aquamist kit.

That's good to know about the seals! I wasn't aware...

That is so awesome you get no knock with 87! Impressive!

Originally posted by hotsam
I run an Aquamist 1s kit and I love it. My install is featured on the Aquamist site here:

http://www.aquamist.co.uk/rescr/gallery/subaru-tm/tm.html

Where did you purchase your kit from?

Matt

Suppliers listed here (http://www.aquamist.co.uk/sl/plist/agent4.html)

Shane Lowrance promises to have the lowest prices on non group buys.

Alternative Performance Products
111 South Ridge Dr.
Cedar Creek, TX 78612
TEL: 512-296-5735
Email: slowrance@austin.rr.com

If you want to talk to a really cool guy and get application and theory from a guy who does it for a living, call Brad Schaffner

Georges Imports Ltd
(SAAB main dealer + technical) Specialist on high flow water
injection systems on large engines
8011 State Line Road
Kansas City
MO 64114
USA
TEL: +1 816 333 6582
FAX: +1 816 361 2096
Email: brad@kcsaab.com
Website: www.kcsaab.com

Matt - empty your pms :)

Send me an e-mail from my profile for more info.

Thanks for the info, Zaxrex!

Pm's emptied...

Originally posted by I am mattjk


Where did you purchase your kit from?

Matt

I purchased it from me (http://www.turbometrics.com). I'm not a nasioc vendor but you can find me on i-club.

Originally posted by zaxrex


DO NOT USE ISOPROPYL ALCOHOL W/ AQUAMIST The pump seals will be degraded by use of iso. Use Methanol at a mixture no more than 50/50 with water.

By the way, water injection is good. I get to run 87 octane gas while commuting to work and still have the TXS 4 knockless. Saves so much money in gas and my milage has increased 20-30%, depending on conditions. In a year with gas savings, I'll have paid for the Aquamist kit.

Now that shows how well water injection works. The fact that you can run crappy 87 octane gas and still not get knock while running a TXS Stg4 (please correct me if im wrong by you saying TXS 4 knockless).

I think I'll just bypass the middle-man and put water in my gas tank :)

Probably running too much water if tuned on 93 and still knockless on 87. Makes for good driveability and efficiency - but not getting the full performance benefit. Not a knock on the decision just a comment. Water does slow down the flame front and ideally you want as little water as you can without knock. You can use less water on higher octane than would be required for 87. That said water saved me at a fuel stop in KS that only sold 87.

I run 94 octane on a 2d with a .6 or .7 jet depending on the environment. I have it set at a low water to fuel ratio since I use a high octane fuel and don't need a lot of water to get knock suppression. On my way out to CO for Pikes Peak - I of course entere the part of the country that thinks corn is fuel. The step down to 91 brought about some knock - I switched to a .8 jet turned up the water pressure - knock was gone - 3 minutes and didn't change any fuel or timing. When I was at the stop that had only 87 went to the .9 jet again no map changes no knock - so it is a good thing to be able to do. I guess my point also is if you want to run 87 - tune WI on your highest octane at a water jet size that just prevents knock - then as you lower octane increase water - this will leave you with the ability to go back to the most power when you want it.

Note on my jet sizes I am using 2d which mirrors the IDC of the FI which provides a fixed water to fuel ratio - rather than 1s which provides a fixed water flow (fixed water flow will be too much water all the time or too little water all the time or too much water on low boost and too little water on high boost). Since with WI my FI never exceeds 75% IDC the jets never are static and flowing their full potential - i.e. .7 flows less water on a 2d than a 1s.

Nice DIY writeup. http://www.frii.com/~maphill/wi.html

so...i am catless and stock ECU management,

can i use this on my car without feeling a loss in power?

i just want to lower EGTs while on long trips. do i have to run the water with boost, or can it just sit there and mist while crusing down the highway...on off switch.

will i feel it when the water kicks in, as in a lag, lull in power delivery, A/C turning on kind of feeling?

how do you determine how much water to inject is a good amount(nozzle size)

thanks

Originally posted by jehcpa
Note on my jet sizes I am using 2d which mirrors the IDC of the FI which provides a fixed water to fuel ratio - rather than 1s which provides a fixed water flow (fixed water flow will be too much water all the time or too little water all the time or too much water on low boost and too little water on high boost). Since with WI my FI never exceeds 75% IDC the jets never are static and flowing their full potential - i.e. .7 flows less water on a 2d than a 1s.

with the 2d system, is the only control over the amount of water delivered determined by the number and size of nozzels you are running, say 1 nozzle(say .4) and then the second one turns on(another .4). so the only control you have is either 0, .4, or .8? you mentioned turning up the pressure in pikes peak, how did you do that? how do you determine the size of the nozzle, pump psi and boost pressure to activate the nozzles...do you just need time ona dyno or some kind of loggin to see what happens and then go from their? i want one of these kits, but do you think i should wait till i can get the utec so i can log and tune instead of blindly throwing ever increasing amounts of water in?

Originally posted by jehcpa
Note on my jet sizes I am using 2d which mirrors the IDC of the FI which provides a fixed water to fuel ratio - rather than 1s which provides a fixed water flow (fixed water flow will be too much water all the time or too little water all the time or too much water on low boost and too little water on high boost). Since with WI my FI never exceeds 75% IDC the jets never are static and flowing their full potential - i.e. .7 flows less water on a 2d than a 1s. Are you sure you have "a fixed water to fuel ratio?" Thanks to its fuel pressure regulator, the WRX's fuel line pressure moves up and down in tandem with boost pressure in order to maintain a constant pressure differential of about 3 bar. The fact that the fuel line/boost pressure differential is constant greatly simplifies the ECU's job - in that it knows the injectors will flow the same amount of fuel for a particular duty cycle, regardless of whether the injector is working against a 1 PSI boost pressure or a 17 PSI boost pressure. Although I haven't looked very closely at your system, I doubt that your water pressure is increasing/decreasing in tandem with boost pressure. And since the amount of water injected probably dependes on the pressure differential between boost and water line pressure, then as boost pressure rises, your nozzle's flow rate decreases, exactly what you don't want to happen.

Jon is probably correct. I do recall seeing another brand of WI where boost pressure was plumbed into a sealed watertank, there by acting like a rising rate system.

To my recollection I have only seen three fuel pressure settings - 1/3, 2/3 and full in DD. Since it is always on full under the conditions that my WI is engaged that shouldn't be an issue.

The pressure differential does have an impact on misting I am not sure that it does on flow. Even if it does we are talking 5 bars vs. boost of .5 to 1.5 bars all the way to redline I have never been short of water. 2 bar differential is significant and all that is required. 3 bars is more than sufficient.

While it may not be a rock solid constant it is within 1% across the band.

Originally posted by Wheeler Bement
so...i am catless and stock ECU management,

can i use this on my car without feeling a loss in power?

i just want to lower EGTs while on long trips. do i have to run the water with boost, or can it just sit there and mist while crusing down the highway...on off switch.

will i feel it when the water kicks in, as in a lag, lull in power delivery, A/C turning on kind of feeling?

how do you determine how much water to inject is a good amount(nozzle size)

thanks

How high are your EGTs on stock management? Unless you are running a CAI they shouldn't be that high for any trip length. I don't think WI would help much but minimal amounts shouldn't hurt.

Originally posted by Wheeler Bement
with the 2d system, is the only control over the amount of water delivered determined by the number and size of nozzels you are running, say 1 nozzle(say .4) and then the second one turns on(another .4). so the only control you have is either 0, .4, or .8? you mentioned turning up the pressure in pikes peak, how did you do that? how do you determine the size of the nozzle, pump psi and boost pressure to activate the nozzles...do you just need time ona dyno or some kind of loggin to see what happens and then go from their? i want one of these kits, but do you think i should wait till i can get the utec so i can log and tune instead of blindly throwing ever increasing amounts of water in?

Standard application of system 2d uses only one jet. The pump has a pressure switch which turns it on and off to maintain constant pressure. Water flow is controlled via a high speed valve that is like a water injector - it fires in the same pattern as the #1 fuel injector.

Originally posted by jehcpa
To my recollection I have only seen three fuel pressure settings - 1/3, 2/3 and full in DD. Since it is always on full under the conditions that my WI is engaged that shouldn't be an issue.



Are you sure that what you are referring to is not fuel pump voltage, and not fuel pressure? It seems to me that with a RRFPR, the fuel pressure would be independant of the the fuel pump voltage. No? It's late, I could most certainly be wrong.:lol:

Steve

I have been wrong before and will be again and Jon has very successfully pointed out those conditions in the past and may have again. I have a new fuel pressure gauge so when I install it I should know more exactly what I am talking about.

Originally posted by jehcpa
To my recollection I have only seen three fuel pressure settings - 1/3, 2/3 and full in DD. Since it is always on full under the conditions that my WI is engaged that shouldn't be an issue.

The pressure differential does have an impact on misting I am not sure that it does on flow. Even if it does we are talking 5 bars vs. boost of .5 to 1.5 bars all the way to redline I have never been short of water. 2 bar differential is significant and all that is required. 3 bars is more than sufficient.

While it may not be a rock solid constant it is within 1% across the band. The 1/3, 2/3, full is how the WRX ECU drives its fuel pump, not its fuel injectors. There appears to be no chance you're achieving a fixed (or anything close to it) fuel/water mixture.

On this I must disagree with you Jon - my WI system also maintains a constant line pressure (through a pressure swtich set below the max potential of the pump) the increase in pressure within the induction track would just increase the pressure from the pump to maintain 5 bars in the WI line. So both fuel and water maintain pressures through their respective jets and therefore consistent flows at any given IDC level.

I have been running (a home made) water injection for a half a year and my experience is if you run too much water, you loose power, too little and it will not do any good. My car is stock and it runs better with consistence performance with water injection. Some one said don't expect power gain, I would say do expect power gain if you 'tune' your water just right. Especially during hard run or on a hot day.

Originally posted by jehcpa
Probably running too much water if tuned on 93 and still knockless on 87. Makes for good driveability and efficiency - but not getting the full performance benefit. Not a knock on the decision just a comment. Water does slow down the flame front and ideally you want as little water as you can without knock. You can use less water on higher octane than would be required for 87. That said water saved me at a fuel stop in KS that only sold 87.

I have my mix of water/methanol. Pour some out onto the ground and you can light it :D . Flame front should be a bit faster than with just water. This way I can lean out the fuel a bit down low to make use of the meth. Then at higher rpms, just bring the fuel back up.

Besides, cruising around here on I(s) 495/95/395/66/270, I'm not in boost all that much and only need 87 for "transportation mode".

Yeah I remember the days I was in Stafford and used to "drive" to the beltway - not much use for power then.

I guess my point was going from 87 to 93 you could probably use less water mix when using 93 octane. Not that you shouldn't use 87 or run less water mix when you are.

Originally posted by mlambert
Jeff from orlando why dont you just come out with your amazing info like you did in the bragging forum..

Lemme guess, this week youre pushing 400whp on 440cc injectors and you put koolaid mix in your water tank?

No not quite, but i got a laugh from your sarcasm. I did however install a greddy evo catback, and man the car hauls! I've also upped the boost to around 20psi, and soon i will attempt to run 22psi. Probably not for another week or two though. Coolaid is not in my plans for the future, washer fluid is working great for me. Its amazing how much fuel is saved when you arn't wasting it JUST FOR THE SAKE OF COOLING, which is isnt even good at doing in the first place.

Jeff

Plus I suspect the request for PM did not have to do with bragging or power claims, but rather to contact about other matters Jeff didn't think would be appropriate to put on the public board without a different status.

In the next month or two i am probably going to a dyno, just as soon as i get all the kinks worked out on the car. Im toying with my adjustable wastegate and boost hoses and such, trying to make more "up-top" boost. I am also progressing towards 22psi on 93 octane, while maintaining a decent afr and idc's. I am getting close now to the maximum the fuel system will go, but i still have some headroom for a leaner mixture past 5500rpms. Even with the same values of -8.3 in my maps, if it reads around 12.5:1, it tapers to "rich" by around 5500rpms. I would like to see at least an even 12:1, if not leaner. That should save a little bit of fuel.

Last weekend i installed a 1gph nozzle before the intercooler, inside my silicon y-pipe, at the spot right before it splits into the 2 directions. It feels like the car doesn't suffer so much from "heatsoak' anymore, although it is still there somewhat. Injecting this small amount of extra fluid should allow me to run a little leaner while still maintaing just as cool cylinders as before. When I get the car pushing the right amount of boost and everything is dialed in i will dyno it at first chance. I am not making any promises but im expecting at least 320whp, possibly as much as 350whp(that is my goal) and torque should be around 290-300 at the wheels as well.

The greddy evo made a huge impact in the car, seeing as i was basically a txs stage 4, making that much power with a crappy stock cat-back was nice, but removing that bottleneck really opened up the doors and the car simply hauls royal @$$ now. It sounds awesome too, nice and low and not too loud.

Anyhow, thats my input for now. There are 3 types of people when it comes to water injection...

Those who have it and love it

those who dont have it but believe in it

those who dont have it, dont believe in it, and show discontent because they are wasting money on race-fuel every weekend. :lol:

Peace

Jeff

Originally posted by jehcpa
On this I must disagree with you Jon - my WI system also maintains a constant line pressure (through a pressure swtich set below the max potential of the pump) the increase in pressure within the induction track would just increase the pressure from the pump to maintain 5 bars in the WI line. So both fuel and water maintain pressures through their respective jets and therefore consistent flows at any given IDC level. It is the pressure difference at the nozzle which determines whether and how much fluid will flow. Reduce the difference and you reduce the flow. On the fuel side, the pressure difference is always maintained at ~3 bar by the fuel pressure regulator. If you were to run 2 bar of boost, the pressure in the fuel line would rise to ~5 bar in order to maintain the desired 3 bar difference.

Jon - I agree there is a need to be as accurate as possible in our posts - however there is discussing things in a reasonable manner with reasonable statements and then there is nit picking accuracy to tolerance levels that just are not helpful, beneficial or constructive.

I acknowledged earlier that constant water to fuel was not exact - but we don't live in laboratories either.

And as I said when the pressure is 7 bars vs. boost of from .5 to 1.5 it matters a lot less at the jet than it does for a fuel system at 3 bars - where it matters a lot more.

The mirroring of the IDC - which it does do exactly. Is still a much better management of water flow to fuel to boost than a static running of a jet.

If we really want to nit pick this topic - then lets get really deep into it:

We are using nylon hose - so when the WI is engaged the hose is stretched and the water flowed initially is less.

The temperature of the water determines how much heat it will absorb when injected - so really what we need is a jet driver that will inject more water when the water is warm and less when the water is cold.

The boiling point of water is different at different pressures - so we also need a system that predicts the pressures that the water will go through the manifold, then the decreased pressure during the induction stroke, then the reincrease to compression. Got an algorithm handy to calculate the ideal water to inject at the IC opening? how about the IC outlet? behind the TB?

The jets themselves are not even exact in how much water is flowed through them at a given pressure.

We can't nitpick everything to death - we would never get a car built.

And here is how much it matters if at one second I am at 16% water to fuel and the next I am at 12% water to fuel.

We are using water to replace the excess fuel being dumped to cool the cylinder and charge. Water has 6 times the latent heat of gas. When leaning AFR from 10:1 to 12.5:1 - the amount of water to remaining fuel necessary to replace the latent heat lost from lowering the fuel is 3%. Now this doesn't take into account the higher charge temperature - so let's double the water to fuel in order to account for the highest charged temperatures experienced - we are at 6%. If I am running around 10% water to fuel my total mixture is 1 part water to 10 parts fuel to 125 parts air - in weight. Unless you are trying to tell me that a pressure differential of 6.5 bars falling to 5.5 bars differential is going to cut my water flow in half - I really don't care. I still am running an optimum fuel ratio, still not getting knock and if anything the lower amount of water is allows a faster flame during higer boosts and higher rpms - net, net all is good.

I will acknowledge again OK it may not be "constant" fuel to water, but it is close enough and it is closer than any method that does not have a stand alone 3D WI driver. Tuning with WI that mirrors IDCs does increase the amount of water as boost increases - which is not the same as the amount of water decreasing in absolute terms (and even more so in terms of water to fuel) like in a static system. Now someone with a lot more money than me can create a laboratory system to exactly match the fuel to water - but I don't have a NASA budget and they would still have a tolerance. And you would still say that it wasn't constant to 24 significant digits.

Heck I only spent $500 and have a more powerful, safer and more efficient tune with a system that does at least start to account for the increases in MAP and fuel - at least 95% of those increases. I will leave the 5% to the extra margin I have by not running 6% water to fuel and run right around the neighborhood of 12%.

Originally posted by jehcpa
And as I said when the pressure is 7 bars vs. boost of from .5 to 1.5 it matters a lot less at the jet than it does for a fuel system at 3 bars - where it matters a lot more.

The mirroring of the IDC - which it does do exactly. Is still a much better management of water flow to fuel to boost than a static running of a jet.Actually you said earlier your water injection line pressure was 5 bars. And the pressure differential does matter, more than you think. It appears E.R.L. has no web presence. But here's a page of misting nozzles that demonstrates what 10 PSI differential can mean in terms of flow rate:
http://service.spray.com/catalog/pdfeng/catalog60B/fine.pdf

And you believe that mirroring the injector duty cycle for one cylinder is a good thing. I do not. Let's consider the simplified case of monitoring cylinder 1 and that it is operating at 75% IDC. If one is to believe the WRX service manual, the ECU times the injector pulse so that it ends at the moment the intake valve opens. So cylinder #1 gets a nice wet mixture, since the water injector has been open for the previous 540 crank angle degrees (75%). But how about poor cylinder #3, the next one in the firing order? For him, the water injection system has been shut down for the most recent 180 crank angle degrees (25%). Does #3 get as much water as #1? I don't think so.

Their site:
http://www.aquamist.co.uk

Unfortunately they do not have a spec list similar to your (yet again another good resource) link.

With a pressure accumulator the pump is capable of delivering 10 bars. I have mine set at right around 7 bars. The switch comes from the factory set at 5 bars. I used 5 bars originally because that is how it arrives in people's hands and my public posts are meant for general consumption and since the manifold pressure doesn't matter at that pressure it definitely doesn't matter at mine.

Also note that proper jet sizing is to match jet/pressure potential to injector potential for the targetted water to fuel ratio.

The gph drop from 100psi to 60psi looks to fluctuate around 30%. Though obviously not linear but lets just say 20 psi of boost drops flow by as much as 15%. Even from 60psi to 40psi the ~20% drops don't concern me with the margin of my settings.

If I am at 10% water to fuel I drop to 8.5% at max boost still safe. If I am at 15% I drop to 12.75% much more than safe.

And since turbo engines are hardest to tune during boost transition (boost increasing faster than the ECU climbs up its map which is why it is run so rich in that area to stay ahead of the boost climb) the relative extra water earlier only helps and isn't needed as much later. The more you reinforce the relative drop in water to fuel (because in absolute terms more water is still being injected at higher boost) the more I like it.

The high speed valve drives off of injector #1. But the jet is in front of the throttle body. The length of the intake and runners is more than sufficient to fully mix the mist with the intake charge. I am not injecting at the valve or even behind the TB. This system has been used with too much empirical success in all forms to be able to claim it is starving a cylinder.

I am running leaner than 12.5:1 AFR at ~17-18 psi (<815*C EGTs) - I don't think it is my 94 octane that has it knock free on a 93* 95% relative humidity afternoon. Last compression test and magnified visual check of the spark plugs shows no signs of detonation.

The exact situation you describe is worse running a static jet - in that situation fuel is going up as much as 75% as boost builds - that means water to fuel is dropping quickly to 57% its original relationship. Someone running static is running too much water early, too little water late - if they are lucky they have it just right as boost peaks. If you have it right earlier you are making the water shortage under high boost worse. If you have it right later you are flooding the engine during transition.

Wow this has turned into quite the topic. Not to mention that this thread is very informative. But I still would like to know more. Especially from people who already have it. Possibly your setup and also what kinda 1/4 miles and/or dyno numbers you have.

PC -

I bracket raced for years so I understand the appeal of going to a local track on Thursday night and setting it off for 1320 - but I just can't buy into dragging an AWD, lose too much in the drive train. You have to do way too much work on the engine to get comparable results of a good RWD car. However, since 1/4 mile is so much to so many and understanding the appeal of it I guess at some point I should at least try to learn to launch this thing and see the results. Sorry I digress - no idea on 1/4 mile in the REX but other WI threads may have mentioned. My prior experience with WI on a high compression motors would indicate that on an optimized system you can drop trap times by anywhere from 2-5% depending on what you are working with before WI.

Another reason for 1/4 mile's is I really hate throwing out dyno numbers. I need to develop a better nonWI set of maps and compare them in as close a manner as possible from optimum nonWI tune to optimum WI tune. Anyway with all the usual disclaimers regarding dynos - on a warm and humid day the last dyno run I had was 235hp, 238tq at the wheels with a catless SS header back exhaust (2 1/2" not 3"), XEDE and WI. Since then I have done some more road tuning and DD road dyno and would guess that I have probably added another 3-5 based on what those are showing run to run. I am still on the stock turbo and there just isn't much more I am going to be able to out of it since with a fully open solenoid pressure is diving right as it is trying to reach new hp peaks. It is close to time to get a new turbo and 3" exhaust, but with WI there won't be any new injectors, fuel rails, regulators etc. So I will recover $1000 on not buying all that junk when I buy my turbo - 100% return on the WI kit not bad.

Here is a link that may give you a rough idea of what to expect with and without WI on any given setup:

http://not2fast.wryday.com/turbo/glossary/turbo_calc.shtml

Here's my question:

What about using the UTEC's flat shifting or using Launch control in fuel cut mode?

I'm curious to know cause I routinely see 8-11 psi during shifts. If WI would be active during that time, could the excess water vapors hinder or even damage the engine?

Originally posted by jehcpa

The high speed valve drives off of injector #1. But the jet is in front of the throttle body. The length of the intake and runners is more than sufficient to fully mix the mist with the intake charge. I am not injecting at the valve or even behind the TB. This system has been used with too much empirical success in all forms to be able to claim it is starving a cylinder.

I'm not trying to inflame things here, but when I was first setting up my STATIC water injection, I was playing around with the different nozzels and trip in points. At times I would have way too much water (before I started mixing it with meth). I taped my nozzel on the underside of the IC outlet collar and when I over watered, I would constantly get simultanesous cyls 1&2 missfires. I take that to mean that 1&2 get more water that 3&4.

That migh just be my nozzel placement and could change with it spraying from the top down, I don't know.

The runners do mix the water well, but I think that there is always going to be differences in water content injested into each cyl. This might only be in the 24 sig digs that you were talking about earlier, but I thopught I would share my over watering experience and that cyls 1&2 always missfired and never 3 or 4.

Originally posted by jehcpa
I am running leaner than 12.5:1 AFR at ~17-18 psi What leads you to believe this? I hope you're not relying on an O2 sensor. Have I ever mentioned that running water or water/alcohol injection invalidates O2 sensor (even with a wideband) measurements of A/F ratio? Well, if not, it does.

That would be news to me - do tell how it invalidates 02 readings. Water injection that is - I don't run a mix as my windshield can attest to.

Keep in mind the only thing that has changed is that less than 1% of the exhaust is made of steam. I have never heard of any condition in which steam would invalidate an 02.

If this is the case you need to start a new thread for the benefit of the members since about 50 people on these boards alone are tuning WI based on AFR - including a recent tune by TXS.

Originally posted by zaxrex


I'm not trying to inflame things here, but when I was first setting up my STATIC water injection, I was playing around with the different nozzels and trip in points. At times I would have way too much water (before I started mixing it with meth). I taped my nozzel on the underside of the IC outlet collar and when I over watered, I would constantly get simultanesous cyls 1&2 missfires. I take that to mean that 1&2 get more water that 3&4.

That migh just be my nozzel placement and could change with it spraying from the top down, I don't know.

The runners do mix the water well, but I think that there is always going to be differences in water content injested into each cyl. This might only be in the 24 sig digs that you were talking about earlier, but I thopught I would share my over watering experience and that cyls 1&2 always missfired and never 3 or 4.

That situation sounds really strange and would take some diagnosis - but I would not conclude that a static jet in front of the throttle body was some how sending more water to certain cylinders unless it was also sending more air. All four of my spark plugs look almost identical when I pull them - how did your plugs look last you pulled them? Could have been the gap made those more sensitive to misfiring or injector volume differences or... like I said more diagnosis would really be necessary.

All four of my cylinders are experiencing similar conditions based on spark plugs and compression tests.

A first blush look at the cyls 1&2 missfire would lead me to suggest your having a problem with spray droplet distribution in the manifold.

The manifold on the WRX is designed as a dry manifold and little consideration was given to what would happen to liquid droplets suspended in the air flow. (unlike a manifold designed for a carburated or single point FI system where it is a given that liquid droplets will traverse the manifold in the air flow)

Sounds like your nozzle is creating a too course water spray under high discharge conditions and the larger water droplets are not successfully making the relatively sharp turn to the near (rear) cylinders. Result most of the large droplets are taking the front two runners which probably have a better ability to carry suspended water droplets. ( or the water is getting centrifuged out and slamming into the front wall of the intake manifold and moving as a liquid film on the surface of the 1&2 manifold runners).

That is one of the advantages of multiple water injection nozzles, it allows you to give a significant portion of the water spray more time to evaporate.


As far as the wideband O2 sensor and the water spray, I would also like to hear how it could be effected, by water injection.

The normal combustion of hydrocarbon fuels produces three major exhaust products, inert nitrogen, C02, and H2O.

If you burn 106 gm of C8 H10 (octane) with 1461 gm of air you will get about 1019 gm N + 352 gm CO2 + 90 gm H2O. So your exhaust in a normal gasoline engine with gasoline only as the fuel will consist of about 5.7% water vapor by weight. This does not include the water weight of the atmospheric water inducted with the air.

To put that a different way, for every gallon of fuel burned you get at least .85 gallon of water in the exhaust.

Since most water injection systems seldom exceed 2:1 ratio water to fuel, you wil only be increasing the water by a few percent of the total exhaust volume. The Alcohol will bias the readings a bit, because it is just a different source of fuel, that has a lower than average Ox requirement, but if that is accounted for the O2 sensor numbers should be valid.

Larry

Originally posted by hotrod
A first blush look at the cyls 1&2 missfire would lead me to suggest your having a problem with spray droplet distribution in the manifold.

The manifold on the WRX is designed as a dry manifold and little consideration was given to what would happen to liquid droplets suspended in the air flow. (unlike a manifold designed for a carburated or single point FI system where it is a given that liquid droplets will traverse the manifold in the air flow)

Sounds like your nozzle is creating a too course water spray under high discharge conditions and the larger water droplets are not successfully making the relatively sharp turn to the near (rear) cylinders. Result most of the large droplets are taking the front two runners which probably have a better ability to carry suspended water droplets. ( or the water is getting centrifuged out and slamming into the front wall of the intake manifold and moving as a liquid film on the surface of the 1&2 manifold runners).

That is one of the advantages of multiple water injection nozzles, it allows you to give a significant portion of the water spray more time to evaporate.

Mostly a reasonable explanation -

The jet does not produce consistent size droplets but it does atomize at any pressure differential over 2 bars. The different size water droplets will accelerate at different rates in the intake charge but none of them would be of a size to create the condition you lay out. (assumes aquamist jets)

Fortunately a very small portion of the water injected will actually evaporate in the intake and any that does is pretty much useless since the primary cooling benefit of WI is water's latent heat of evaporation. The water injected will not evaporate past 100% relative humidity and a lot of heat will be absorbed first heating the water then evaporating the water that does evaporate and would thus result in some of the evaporated water condensing.

That said if too much water is being injected the atomized droplets may merge into droplets so larger that they would tend to go into the front runners. So that is a valid explanation but it would be because of too much water rather than a condition that exists with any water injection amount.

I assume the manifolds on the WRC cars are designed similarly as dry - and with the exception of the two french teams all use aquamist jets.

As far as the wideband O2 sensor and the water spray, I would also like to hear how it could be effected, by water injection.

The normal combustion of hydrocarbon fuels produces three major exhaust products, inert nitrogen, C02, and H2O.

If you burn 106 gm of C8 H10 (octane) with 1461 gm of air you will get about 1019 gm N + 352 gm CO2 + 90 gm H2O. So your exhaust in a normal gasoline engine with gasoline only as the fuel will consist of about 5.7% water vapor by weight. This does not include the water weight of the atmospheric water inducted with the air. Since most water injection systems seldom exceed 2:1 ratio water to fuel, you wil only be increasing the water by a few percent of the total exhaust volume. The Alcohol will bias the readings a bit, because it is just a different source of fuel, that has a lower than average Ox requirement, but if that is accounted for the O2 sensor numbers should be valid.

Larry

Succinctly put my exact wonderment how the introduction of less than 1% of additional steam would invalidate an 02 sensor.

Since I only use water - I wasn't going to get into the fact that I realize the stoich of methanol is different than gasoline and that most readers actually measure the calibrated lambda and then translate int the calibrated AFR - but as I tried to raise earlier, then you are worrying about the diminimus - less than 0.5% of the intake charge would be methanol. I will be glad to calculate (or more likely get someone else to calculate) the average lambda of 20 parts fuel to 1 part methanol and put the appropriate AFR calibration in my reader. What it might be 14.4 instead of 14.6? Not exactly an invalidation.

Either way I can be highly confident that even the stock O2 reading 12.8 is in fact higher than 12.5 with WI or WMI.

Fortunately a very small portion of the water injected will actually evaporate in the intake and any that does is pretty much useless since the primary cooling benefit of WI is water's latent heat of evaporation.


Don't underestimate the value of the cooling of the intake charge before it gets to the intake valve. That is one of the mechanisms that allows water injection to improve performance, by increasing the engines VE.

Water injection does many things not just cooling the combustion process.

Larry

By no means do I underestimate the density it has the potential to add to the charge, but in a high compression its most impressive work is done in-cylinder during the induction stroke cooling the cylinder and during the compression stroke by suppressing knock.

Adding density to the charge does occur but if that was the only benefit the gains achieved would be much less. A properly designed, maintained and externally sprayed IC will do more for the pre-induction charge than WI could hope to achieve - it is limited by air to air cannnot reduce beyond ambient without sub-ambient spraying and relative humidity limits the ability to get there internally.

External assistance with much larger volumes of external air and heat transfer are best for hot charges.

Originally posted by zaxrex


DO NOT USE ISOPROPYL ALCOHOL W/ AQUAMIST The pump seals will be degraded by use of iso. Use Methanol at a mixture no more than 50/50 with water.

By the way, water injection is good. I get to run 87 octane gas while commuting to work and still have the TXS 4 knockless. Saves so much money in gas and my milage has increased 20-30%, depending on conditions. In a year with gas savings, I'll have paid for the Aquamist kit.

:eek: :eek:

if that was the only benefit the gains achieved would be much less.


Never said it was the only effect, just an important one.

Intake tract cooling can be 30 deg F or more depending on alcohol percentage, below the temp the air leaves the intercooler. That can be worth a 3% power increase independent of detonation supression.

Detonation supression not only is due to cooling during the intake stroke process and cooling the cylinder head, valves etc, but also occurs due to changes in the combustion process itself by modification of the chemistry of the combustion.

Knock occurs during the power stroke not during the compression stroke. Detonation takes place late in the combustion process.

Larry

Originally posted by hotrod
Knock occurs during the power stroke not during the compression stroke. Detonation takes place late in the combustion process.

Detonation may be sensed during combustion - but combustion is started during the compression stroke. How and when combustions starts and how combustion proceeds is determined before TDC of the compression stroke and it is during this period that detoanation is suppressed whether it be by octane, water, methanol or timing.

Originally posted by jehcpa
PC -

I bracket raced for years so I understand the appeal of going to a local track on Thursday night and setting it off for 1320 - but I just can't buy into dragging an AWD, lose too much in the drive train. You have to do way too much work on the engine to get comparable results of a good RWD car. However, since 1/4 mile is so much to so many and understanding the appeal of it I guess at some point I should at least try to learn to launch this thing and see the results. Sorry I digress - no idea on 1/4 mile in the REX but other WI threads may have mentioned. My prior experience with WI on a high compression motors would indicate that on an optimized system you can drop trap times by anywhere from 2-5% depending on what you are working with before WI.

Hello jehcpa,

No need for launching your WRX and abusing your tranny. Heck, stick with a 2.2-2.3 60ft time. Just be consistent like when you bracket race. All you need to find out is how much power your are making before and after runs/adjustments. Just go by your mph. As much as I love 1/4 miling and bracket racing, I don't even care for 1/4 mile times anymore, only trap speeds.

Nice post and lots of info. :cool:

Originally posted by hotrod
As far as the wideband O2 sensor and the water spray, I would also like to hear how it could be effected, by water injection.

The normal combustion of hydrocarbon fuels produces three major exhaust products, inert nitrogen, C02, and H2O.

If you burn 106 gm of C8 H10 (octane) with 1461 gm of air you will get about 1019 gm N + 352 gm CO2 + 90 gm H2O. So your exhaust in a normal gasoline engine with gasoline only as the fuel will consist of about 5.7% water vapor by weight. This does not include the water weight of the atmospheric water inducted with the air.

To put that a different way, for every gallon of fuel burned you get at least .85 gallon of water in the exhaust.

Since most water injection systems seldom exceed 2:1 ratio water to fuel, you wil only be increasing the water by a few percent of the total exhaust volume. The Alcohol will bias the readings a bit, because it is just a different source of fuel, that has a lower than average Ox requirement, but if that is accounted for the O2 sensor numbers should be valid.The problem is that all the water injected into the intake will also be added to the exhaust. One effect of this extra H2O in the exhaust is a reducuction of the O2 partial pressure in the exhaust from what it would have been if no water had been injected. The ZrO2 sensor cells of an A/F ratio sensor are designed to respond only to the O2 partial pressure. The end result is that, with water injection, any A/F meter will report an A/F ratio richer than was actually present in the intake.

Thanks Jon!

The end result is that, with water injection, any A/F meter will report an A/F ratio richer than was actually present in the intake.

That makes sense, I see what your saying.

Due to the increased exhaust gas volume, as a percentage, the residual oxygen is much less then it would have been in a dry engine.

Hmmm -- need to ponder that for a while and see if there is a simple way to dial out/correct for that effect. The biggest problem I see on first consideration is that most all WI ADI systems do not do a precise job of metering the injected fluid like the fuel injectors do. At any given moment you have a substantial uncertainty about how much extra fluid was added to the combustion process.

Larry

The water being added is diminimus in both weight and volume, I find it highly suspect that my lambda readings are being affected enough to alter the translated AFR by more than 0.05.

Even accepting that AFR may vary slightly as a result it is hardly invalidated. Besides maximum power AFR is not an absolute that is constant accross motors. Whether my actual AFR is 12.4 or 12.6 when a gauge reads 12.5 is not relevant because I am not targetting a particular AFR. I am adjusting fuel at various load points to reach the point of maximum power for that point which is dependent on VE from air filter to exhaust tip. WI permits me to work in any of these ranges without knock. Unfortunately it is very iterative and dyno time costs a lot and analysis from road tuning takes time - but I am getting there.

WI does give up some by slowing the flame front, however the gains from reducing fuel that is being used to cool the cylinder and charge is well worth it. There is no other way that I am aware of that will permit me to eliminate the power robbing costs of fuel dumping for knock suppression.

Originally posted by hotrod
The biggest problem I see on first consideration is that most all WI ADI systems do not do a precise job of metering the injected fluid like the fuel injectors do. At any given moment you have a substantial uncertainty about how much extra fluid was added to the combustion process.

Water volume and hence weight is nowhere near consistent enough to accurately adjust. You could adjust based on the mean specification (which I assume is what most jet manufacturers publish) this could be accounted for in the calibration and translation. The tolerance from mean in application I expect is wider than desired for these measurments. However in tuning an open loop map I do not see the benefit. I do see how this can cause problems with a closed loop system that is targetting lambda or AFR but even then the amount of water involved is probably low enough to not matter to someone that is developing a tune for a road car even in closed loop.

I would be interested in knowing based on a proper calculation what the stoich AFWR is for say ?:6.67:1 and ?:10:1. Or alternatively how to calculate such on my own.

Jon -- Not calling your explanation into question, but what are the quantitative PP O2 differences and what is the magnitude of the resulting change in voltages between non WI and then WI cycles?

Are we talking 1-2% or numbers that are orders of magnitude higher than the precision/accuracy/sensing parameters of the sensor/meter?

Jon - over time you have corrected and clarified my comments, have proven me wrong (as I have readily admitted in those cases) and generally increased my knowledge. And I can not recall a time when you have mispoken to any level of significance. But this AFR invalidation claim continues to bother me - both because of your past contributions to my knowledge and because you haven't provided the usual stunning support that you have in the past.

SAAB does not seem to have any problems running close loop fuel control (which requires lamda accuracy) with WI. We should try to help them out.

from SAAB emmissions work in Automotive Engineer (http://www.aquamist.co.uk/dc/coollinks3/index/rally/saab/press/press.html)
The Lambda sensor is in a closed loop between the engine exhaust and ambient air as inputs, and fuel-metering via the ECU on the output side. Its function is to maintain the air-fuel mixture at the optimized ratio for minimal emissions, when Lambda = 1.

As a side not for all WI decreases NOx emissions at all levels.

Also for those looking for the power increase see the graph of the increase in power they were able to otain with WI.

Hmmm-- I need a bit of elaboration on this comment, not entirely sure I understand what your looking for.


I would be interested in knowing based on a proper calculation what the stoich AFWR is for say ?:6.67:1 and ?:10:1. Or alternatively how to calculate such on my own.

The stoich AF ratio should be completely independent of the amount of supplimentary water injected. If the user has an alcohol mix that will complicate it a bit as it will add to the fuel ratio, and has a substantially different oxygen requirement.

If I understand what you asking is how do you figure the stoich of a Air Fuel water alcohol system, it should be pretty straight forward if you know the injection rates, fuel, and alcohol type.

For the basis of discussion, the serious high power users of water injection are the unlimited air racers and the unlimited hydroplanes that use old wwII vintage air craft engines.

They generally run a mix of 50% 50% water methanol as the ADI fluid and inject it at a rate of .5 lb of ADI fluid to each 1 lb of fuel.

That would give you (ignoring for a moment the differences in fluid density) a ratio of FWM (fuel water methanol) of about 4:1:1 (by volume). Assuming they are running a max power lean fuel air ratio of about 13.5:1, then you would have a total ratio by mass of about 54 parts air, 4 parts fuel, 1 part water and 1 part methanol.

Unfortunately the fuel and water alcohol mix are both metered by volume flow rather than mass flow so that would have to be accounted for.

You should be able to figure out the proper true ratio by computing the oxygen requirement for each fuel componet, (e.g. gasoline, methanol) and useing the mass ratio of oxygen in air of about 23% figure out the lbs of air required to meet what ever total fuel:air ratio you had in mind.

Larry

Originally posted by zaxrex
Jon -- Not calling your explanation into question, but what are the quantitative PP O2 differences and what is the magnitude of the resulting change in voltages between non WI and then WI cycles?

Are we talking 1-2% or numbers that are orders of magnitude higher than the precision/accuracy/sensing parameters of the sensor/meter? Well, I don't have much beyond my suppositions. Here's an excerpt from Dr. Heywood's Internal Combustion Engine Fundamentals whose text from the previous page ended: "Dry exhaust gas composition data, as a function of the fuel/air equivalence ratio, for several different multi- and single-cylinder automotive spark-ignition engines over a range of engine speeds and loads are shown in Fig. 4-20. The fuel compositions (gasolines and isooctane) had H/C ratios ranging from 2.0 to 2.25 Exhaust gas composition is substantially ..."
http://Jon.in.CT.home.att.net/ExhaustGasOxygen.gif
I know, it's way too busy. But lean is left and rich is right and the O2 line descends rapidly as it approaches stoichiometric (equivalence ratio = 1) and then tapers its descent very gradually in the rich region. As you can see, a very small change in O2 in the richer-than-stoichiometric region represents a large change in A/F ratio.

My understanding of how ZrO2 sensors work is based primarily on this paper: http://vehicle.me.berkeley.edu/~markw/efi/SAE930352/. Sadly, much was omitted from that transcription. I, given my meager understanding of chemistry, can't decide whether CO2 in the exhaust contributes to O2 partial pressure (any help here?). BTW, I believe the Ford SAE paper was prepared primarily to rebut an SAE paper submitted by some EPA bozos that had appeared the previous year announcing the startling conclusion that automotive oxygen sensors responded to hydrogen, not oxygen. The 'hydrogen' paper may be found at http://vehicle.me.berkeley.edu/~markw/efi/SAE920289/. Ignore the EPA paper, except when someone tries to tell you that O2 sensors read hydrogen, not oxygen.

I also need to note that Saab's water injection kit for Ecopower engine "... operates only during period of full-throttle accleration and then again at road speed above 220 km/h." In other words, the Saab water injection system only runs when the A/F system is in open-loop and ignoring its oxygen sensor.

BTW, I posted this message (http://autos.groups.yahoo.com/group/oz-diy-wb/message/4113) in the oz-diy-wb message forum, in response to a post by Peter Gargano, guru:From: "jon_in_connecticut"
Date: Thu Aug 21, 2003 6:45 pm
Subject: Re: Watr Injection

Is there any update on the issue of water injection and how it
affects a meter's reported air/fuel ratio?

--- In oz-diy-wb@yahoogroups.com, Peter Gargano <peter@t...> wrote:
> Can anyone help out specifically on this?
>
> AH wrote:
> > Will the accuracy of the techedge 1.0 output be affected if I am
> > running a water injection that is injecting 20% of the total
fuel flow?
>
> Yes, there is a dilution effect as there is more inert gas.
> We don't have any specific figures we can give you yet, but
> there are people who can, and we'll put up more info on our web
> site when we can ourselves ..
>
> I believe it's a little more complicated than giving you a simple
> conversion equation as it's the end gas volumes that you must
> factor into any dilution equation, not the input flows/volumes.
>
> Any WI will affect any AFR meter, so the effect will be similar
> for version 1.0, 1.1/1.5 and also version 2.0 (or a Motec, etc.).
> With version 2.0 we have the opportunity to do real time
calculations
> of actual AFR relative to the amount of WI you're using (but we
> have not put this on our 2.0 agenda - yet!).
>
> Peter.There has been no response to my message nor any website update, as far as I can tell.

I also posed this question at the LM-1 forum, http://autos.groups.yahoo.com/group/InnovateLM1/messages, but whoever moderates that forum must work 9-5, M-F because my question is not yet visible there.

Originally posted by Jon [in CT]
I also need to note that Saab's water injection kit for Ecopower engine "... operates only during period of full-throttle accleration and then again at road speed above 220 km/h." In other words, the Saab water injection system only runs when the A/F system is in open-loop and ignoring its oxygen sensor.

Your conclusion is different than mine. I take the statement to be that it full-throttle and above 220kph are independent states for operation and not joint.

I also take the totality of the statements around the quote to mean that for these tests they were running closed loop fuel control all the time holding lambda at 1.0. I see no indication that they are running open loop ever - or else the references to close loop are pointless. I will inquire further for clarification from the author or find out from those who performed the testing.

Originally posted by Jon [in CT]
As you can see, a very small change in O2 in the richer-than-stoichiometric region represents a large change in A/F ratio.

Am I to take this to mean that the injection of water displaces sufficient air during induction to reduce the O2 content of the exhaust to cause a sensor to report richer conditions than existed in the mixture?

At a fraction of a percent I find this difficult to believe.

I do not know enough about mole fractions to be able to see that a less than 1% reduction in the amount of air in the intake would reduce exhaust O2 sufficiently to move me to a significantly richer condition. I know enough not to take a visual representation as answering the question - scale matters and I do not know enough about the relative presentation of the scales to conclude one way or the other. Though the visual presentation may appear to support a small drop (as presented graphically) in O2 results in a richer condition I would rather see the underlying data so that I could calculate percentages as I do not trust graphs and scales.

Originally posted by hotrod
Hmmm-- I need a bit of elaboration on this comment, not entirely sure I understand what your looking for.

I am not sure I do either.

I am trying to put my arms around how much the sensor would be impacted by the additional steam in the exhaust through the introduction of water into the induction.

If my actual AFR is 12.5 - what would a WBO2 calibrated for "dry charge" report the AFR as when I am injecting less than 1% water into the intake charge? Is there a way to get to such an answer through chemical equations or would a controlled test be needed? Would the reported AFR be 12.4 or would it be 11. Clearly the first answer doesn't matter to me but the second one would mean a lot.

You could probably compute it based on steam tables etc. I'm at work so a little reference resource deprived.


I think the easiest way to do it would be like you suggested a simple test.

Reduce the boost to a pressure that eliminates knock and shut off the WI and measure your AFR. Unless I'm missing something your AFR should be the same at a lower boost level.

With the possible exception of at high boost if you were pushing the turbo out of its efficiency range you'd be loosing some air flow due to intake charge heating. But I think what you're really looking for is relative data, not lab quality test data.

Like you mentioned above for basic tuning considerations it should not be a big deal if you miss an absolute AFR number by a few percent.

Max power mixtures range from about 12.5 on the rich side to about 14.2 on the lean side. For most engines there is only about 1% power differential to be had by hitting the absolute max power AFR somewhere near 13.3.

In short the highly paid engineering types don't bother trying to get that last 1% because it is too risky, much safer to settle on a max power rich setting near 12.5 and call it a day.

In the case of WI, they have tested leaner mixtures (which are now much safer) and continued to gain power up to at least 13.5:1 AFR.

Most of the big boys in air racing with WI ADI systems run near the max power lean side around 13.2 - 13.5:1 as I understand it. Of course for them there is also a weight penalty for carrying any more fuel than necessary, so running a lean mixture for lower fuel consumption is a distinct advantage regardless of power produced.

Larry

Originally posted by jehcpa
If my actual AFR is 12.5 - what would a WBO2 calibrated for "dry charge" report the AFR as when I am injecting less than 1% water into the intake chargeIf you're only "injecting less than 1% water into the intake charge," then you should have no interest in this discussion. This thread is for folks interested in water injection.

Not to sidetrack the current discussion, but can anyone explain to me why I would want to put the nozzle in the intercooler (I'm running a TMIC) and not in the throttle body right before the throttle plate?

Originally posted by hotrod
You could probably compute it based on steam tables etc. I'm at work so a little reference resource deprived.

I'm enjoying your discussion...

Here's a mollier diagram for ya:
http://alan.net/austin/Subaru/AirPsychrometricChart.bmp

Here's a steam table for saturated steam:
http://alan.net/austin/Subaru/stmtbl1.jpg
http://alan.net/austin/Subaru/stmtbl2.jpg

I wourld prefer a location for the nozzle in the outlet of the intercooler over the throttle body for several reasons.

If you place the nozzle right infront of the throttle plate, the water mist will not have much time to mix with the whole air stream, before it gets to the throttle plate.

Some of it will also come out of suspension and wet the throttle plate possibly giving you eratic water air ratios especially during throttle transistions like during shifting, or brief throttle lifts in corners etc.

Placing it farther back in the intake hose or the outlet of the intercooler will give more mixing time. Due to evaporation the water drops will be smaller and will more be more likely to stay in suspension as they pass the throttle plate and make turns into the manifold.

I also like the idea of not drilling a hole in a very expensive throttle body. A hole in the outlet of the inter cooler is easier to do, and easier to repair if you change your mind later.

My 2C
Larry

Thanks Austin ---- I think??

The part I forgot to mention was the 3 -4 hours of re-reading 30 year old text books to get my mind around that stuff again. :D :D ;)

I'm sure someone else here is much more current on partial pressures of steam than I am.

I'll have to pass for now on solving that problem.

Larry :cool:

My understanding of the operation of the wideband a/f sensors is that it is a feedback loop system that causes the a/f ratio in the test volume of the sensor to go to stochiometric ratio. It accomplishes this by electrically pumping oxygen either out of or IN TO the test volume until the ratio is stochiometric. When the test volume is rich, the pump cell takes oxygen from the OUTSIDE, and pumps it in to consume the extra fuel. Thus, I would think that the residual oxygen in a richer than stochiometric exhaust really shouldn't affect the accuracy of the measurement substantially. If there is a different kind of fuel, ie, methanol, this will change the stochiometric point, and should change the reading of air/fuel ratio.

Here is a link to an explanation of the a/f sensor operation:

wideband sensor tech info (http://techedge.com.au/vehicle/wbo2/wbntk.htm)

Hope this helps

Mark

Originally posted by Jon [in CT]
If you're only "injecting less than 1% water into the intake charge," then you should have no interest in this discussion. This thread is for folks interested in water injection.

Huh???

An intake charge mixed with fuel to air to water of 125/10/1 is 10% water to fuel. That is about 0.75% of the intake charge. Just how much water do I need to inject before I should be interested in this discussion? 20% fuel to water as you sent to DIY? That is too much for most street applications but if you insist that is only 1.46% of the intake charge.

You originally stated that WI invalidates AFR readings. Those interested in WI would like to know how it has been "invalidated" or of higher probability how large is the rich error in the exhaust with 1% water added to the intake charge? Since you have bypassed the question many times now I can only guess that the "invalidating" of AFR has been withdrawn.

I am going to leave it at this - with the amount of water being injected relative to the total intake charge the impact on the accuracy of a WBO2 is minimal - until I see evidence to the contrary. I think the many others out there using WI with great success will keep working on that assumption as well.