Why water injection is the best mod I've done in years...

[crazyboost]

Thank's a lot, I've read all the thread, and appreciated people sharing their experience.

DSM PERFORMANCE - I believe you would actually tune to a lower air/fuel ratio then 12.5:1 because windshield washer fluid contains methanol which has a stoich ratio of around 6.4:1 So the stoich ratio of the gas methanol mix would be lower then 14.7:1. I'm sure you could a lil bit more boost depending on how good your gas is, but here is some more info for you from a previous post about afr's and water/meth injection:

Quote:
Stoich Values:
Gasoline = 14.7
Methanol = 6.40

Weight:
Water = 8.33lbs/gal
Methanol = 6.59lbs/gal

Specific Gravity:
Gasoline = .737
Methanol = .791

Specific Gravity Differences:
Gasoline = 1.07 more dense than methanol
methanol = .93 less dense than gasoline

Equation to get Percent Methanol to Water Injection by Mass:
(Methanol Weight per Gallon)(Volume of Methanol in Mixture) = Methanol Total Mass
(Water Weight per Gallon)(Volume of Water in Mixture) = Water Total Mass

(Methanol Total Mass)(100)
(Meth Total Mass) + (Water Total Mass) = Methanol Percent by Mass

Equation to get Actual Percent of Methanol Injected:
(Methanol Percent by Mass)(Total Percent of Water Injection Mixture) = Methanol Percent Injected

Equation to get Mixture Stoich Value:
(Meth % by Mass)(Meth S.G. Difference)(Meth Stoich) + (gas %)(gas S.G. Difference)(gas Stoich) = Mixture Stoich Value

Equation to get REAL AFR:
(Mixture Stoich Value/WBO2 reading) x 14.7 = REAL AFR


Example:
75% meth injection to 25% water by volume at 20% of fuel delivery and a WBO2 reading of 11.2:1AFR.

Percent Methanol in Water Injection Mixture by Mass:
(6.59lbs/gal)(.75) = 4.9425

(4.9425)(100)
4.9425 + 2.0825 = 70% Methanol by Mass of 20% Water/Methanol Injected

Equation to get Actual Percent of Methanol Injected:
(70%)(20%) = 14% Methanol injected to Fuel

Mixture Stoich Value: (.14)(.93)(6.4) + (.86)(1.07)(14.7) = 14.36

Real AFR: (11.2/14.36) x 14.7 = 11.47 REAL AFR

 

[compression]

Fuel injector size X 4 cylinders X maximum FI duty cycle = fuel system
size

For my car: 550cc/min X 4 X 95% = 2090cc/min of fuel

A 20% water to fuel ratio would come out to: 418cc/min
(2090*20%=418)

I used a 5mm to cool my intercooler (about 190cc/min)
I used a 6mm to cool pre TB (about 230cc/min)
190+230=420 which is right where I want to be.
Hope this helps,
Jonathan

so im 650x4x95%=2470 2470x20%=494, my question is, i am going to make my own kit. a boost controlled switch with a pump and the existing windshield sprayer tank, and use two 240cc injectors, i will set the switch to like 22psi (i run 23-24 with stock smic) and i can put 1 injector before the intercooler and 1 before the tb, but like you said , i will start with 1 first , i guess the one before the intercooler. if i am satisfied with the results i will just use that one., now did i miss anything?

 

[Slippi84]

so im 650x4x95%=2470 2470x20%=494, my question is, i am going to make my own kit. a boost controlled switch with a pump and the existing windshield sprayer tank, and use two 240cc injectors, i will set the switch to like 22psi (i run 23-24 with stock smic) and i can put 1 injector before the intercooler and 1 before the tb, but like you said , i will start with 1 first , i guess the one before the intercooler. if i am satisfied with the results i will just use that one., now did i miss anything?

Yeah if you think that the best one to start off with is before the intercooler. Let the intercooler do it's job then inject the water/alky or w/e it is you choose. It was discussed either earlier in this thread or in another that one you will have build up of alky in the intercooler and two and intercooler can only make tha air so cool, so trying to pre cool the air charge before the intercooler is negating the intercooler.

 

[dsm-onster]

Absolutely... Think of it like this. I can get my intake temps to below ambient (outside temps) w/ injection. An itercooler can only lower the intake temps to ambient becasue it is using the ambient temps to lower the intake tems. If I cool the temps to below ambient then send it through an inercooler that is at ambient temperature, then the intercooler is heating up the charge !

Let the intercooler cool down the air as much as it can. Then it takes less injection to achieve the same results.

 

[compression]

ok i have my water injection kit on and installed, i put tap water in there to try it out

should i be ok spraying 240cc's through a nt 4g63 injector powered by an old tank and pump off of a cavalier , its squirts pretty good, but it sprays more of a stream then a mist.

i want to try it out, am i safe for take off?

 

[compression]

well i found some extreme blue windshield washer fluid (methyl alcohol) and epmtied all my water out and put that in there, i cant hear the pump turn on due to my exhaust being loud but i can only imagine its working , i logged it and i got 24 knock max and 90 indc on , get this 30psi well it either spiked 30 and settled to 26 or spooled to 30 and fell to 26 (i think it fell because im only on a mildly ported evo 3 16g.) but it seemed like it was pulling really hard. would this be ok to run at the track? i have a 240cc injector doing all the squirting, and by the equation i should be able to run 494cc's so should i be ok to run a 390cc injector or maybe even a 450cc? and lean it out even more? maybe give it a little more timing? . well so far this is one of the best mods ive done, id like to get some cams to just top it off , as well as a bigger turbo ( sad because i just got my evo3 )

but all seems to work well 24 knock should subside a little when 110 is added right?

all this was done on 93 pump gas

 

[RavenX]

Ive been into water injection for a few weeks now and i am stepping up from water to a meth/water mix. My question is what do you guys think about nitromethane? You know the fuel you use in gasoline powered r/c cars. I would really like to try that but if its too dangerous then forget about it. Just would have been interesting to see what would happen with a mixing that with water and trying it out.

 

[Slippi84]

NEVER DO THAT AGAIN!!!

24 counts of knock is asking for detonation and premature engine failure. most people tune for 3-5 counts of knock. you realise that at 24 counts of knock your timing was being pulled about 8 degrees. Trying to run 30 psi on anything no less a 16g is crazy on pump gass. It may have felt like it was pullin hard but you weren't pullin any harder than if you had it at like 24- 25 psi because at 30 psi on an evo your just shooting hot air unless you have some realllly good cooling(huge fmic water injection ect..)

 

[dsm-onster]

He can do just fine @ 30 psi and pump gas w/ water injection. Properly delivered water injection will raise effective octane leves to around 110. That's race gas octane. I ran 30 psi on stock cams and my 18G which everyone swears is hardly bigger than an EVO3 16G. I recorded 1 degree of timing retard at torque peak.

His water injection setup is flawed:

First, he needs to run straight water as he does not have enough flow from his injector to run a water/alky mix. Running straight alky requires 3 times more flow than straight H2O.

Second, his "nozzle" or injector is not misting properly. This in turn does not give him much of the added benefit of cooling the aircharge before it gets into the combustion chamber (like what an intercooler does, but on a much grander scale).

Third, is his fuel pump flowing enough water/alky? Who knows? It's a chevy cavalier fuel pump and this is dsmtuners.com.

If he plans on running this boost often, I do suggest and intercooler upgrade. Not so much for the cooling, but for the increase in flow.

I will be running my RS60T (60-1) at 30 or so psi w/ pump gas this season, thanx to water injection.

 

[dsm-onster]

Ive been into water injection for a few weeks now and i am stepping up from water to a meth/water mix. My question is what do you guys think about nitromethane? You know the fuel you use in gasoline powered r/c cars. I would really like to try that but if its too dangerous then forget about it. Just would have been interesting to see what would happen with a mixing that with water and trying it out.

You'll have to find out if nitro will cool down the combustion chamber and cool the aircharge. Don't most guys just run straight nitro to get what they want? I think nitro releases mor energy when combusted and of couse is less prone to detonation... I could be entirely wrong. So hopefully someone will chime in and save me .

 

[compression]

NEVER DO THAT AGAIN!!!

24 counts of knock is asking for detonation and premature engine failure. most people tune for 3-5 counts of knock. you realise that at 24 counts of knock your timing was being pulled about 8 degrees. Trying to run 30 psi on anything no less a 16g is crazy on pump gass. It may have felt like it was pullin hard but you weren't pullin any harder than if you had it at like 24- 25 psi because at 30 psi on an evo your just shooting hot air unless you have some realllly good cooling(huge fmic water injection ect..)

right, well on 22psi without water injection i was at 36-43 knock, now ar 24 knock with 30 psi and only 240cc's of injection technically by the equation i can run 494cc's just like was stated, 1 nozzle removed half knock the second took it away, i only have 1 small nozzle with room to go. i say it worked out pretty good.

it was at 30psi mostly mistakenly , as i was at 20psi and i turned the boost up a few turns.

i quickly turned it down to 25-27psi range. it runs fine, after a few races last night though the reducer for my intercooler and uicp had a hole blown in it, so i need a new coupler.


i think with 110 octane at 30psi with a 390cc injector instead of the 240cc , at 29-30 psi (if this can be achieved) i will see some high 11's (hopefully)


i dont think my timing was being pulled quite that far, because i have an afc so when i took fuel out of the top end it automatically uses higher timing maps , so if it didnt knock my timing would have probably peaked somewhere around 26-27* so i believe thats where most of the knock is coming from. but with a larger injection i think i will be ok.

what do you think?



p.s. the pump is flowing more then enough. i had to clamp every hose becaue they blow off when it kicks on even with the injector wide open.

 

[dsm-onster]

p.s. the pump is flowing more then enough. i had to clamp every hose becaue they blow off when it kicks on even with the injector wide open.

That means nothing. I have to tighten my clamps with the finger of God for my water injection setups. Who's to say that the pressure it take s to blow off your hose is enough to blow off mine (that cound be take n the wrong way ). Besides, presure and flow are two totally different tthings.

 

[Slippi84]

I didn't just make up that number of 8 degrees timing retard that's how the stock ecu works. For every 3 counts of knock I think it is it retards the timing 1 degree. Now I understand how the safc puts you in a diffrent timing map and is notoriously known for increasing timing due to the ecu thinkin it's seieng less air and using more agressive timing curves but that's only in the abisence of knock. When knock is present like I said depending on how much knock your seieng is how much timing will be retarded."7. Excessive amounts of ignition retard will cause a loss of power and overheating" The reason your car is pulling harder now is that even though like dsm0onster said your water injection was flawed it still managed to decrease your knock from 34(which by the way I have to look up cause I have never heard of someone not hurting the engine with that much knock) to 24. So simply by the fact that you ran more boost and kept knock down it's running better which is the key and the idea of what water ijection is suposed to do. Problem is you would pull even harder and make even more power if you could figure out why your knockin so much not to mention you would descrease your chances of premature engine failure.

 

[Slippi84]

Now, while the ECU has tables for the amount of fuel it needs to inject, it also has table for how much timing advance it should give you, and tables for how much it should advance timing depending on knock.

For 0-3 counds of knock, the ECU will advance timing. For 4-7 or so counts, it will leave timing where it is, and anything over 7 will result in the ECU bringing the timing down in an attempt to control the detonation. While 2g guys cannot view this knock sum on a logger, it is there, you just have to guess what it is by the behavior of the timing curve.

Now, the timing tables in the ECU, just like the fuel maps, are indexed by airflow and rpm. With a SAFC, this has an added effect. Since a SAFC intercepts that signal from the MAS to the ECU and modifies it, it can change the amount of airflow that the ECU "sees." If you have to correct your SAFC into the positive range, than the ECU will see more airflow Hz than the MAF is outputting, and could change the timing map you are following. The problem with this is, higher airflow levels get less timing advance for safety, and lower airflow levels get more timing advance, because the ECU thinks you are pulling in less air.

By leaning out the SAFC (big injectors, more fuel pressure, race gas) you decrease the amount of airflow that the ECU sees, and therefore you usually will get a bit more timing advance for power. This all assumes you have no knock, and also keep in mind that more timing advance will have an engine a higher propensity to knock.

I have heard of 1g guys with 660 cc/min injectors getting timing advanced as much as 28+ degrees at WOT, because you hve to pull the SAFC correction factors down a lot due to the fact that 660's are 47% bigger than the stock 450's.




That's from the tech guide. So after you went above 7 counts your timing got retarded.

 

[dsm-onster]

Not to be nitpicky but . . .
(1 - 450cc/660cc) * 100% = 31.818181818181818181818181818182%

 

[Slippi84]

Not to be nitpicky but . . .
(1 - 450cc/660cc) * 100% = 31.818181818181818181818181818182%

I didn't write that its' from the tech guide. But good thing you corrected it. EIther way 24 counts of knock equals puled timing

 

[compression]

Now, while the ECU has tables for the amount of fuel it needs to inject, it also has table for how much timing advance it should give you, and tables for how much it should advance timing depending on knock.

For 0-3 counds of knock, the ECU will advance timing. For 4-7 or so counts, it will leave timing where it is, and anything over 7 will result in the ECU bringing the timing down in an attempt to control the detonation. While 2g guys cannot view this knock sum on a logger, it is there, you just have to guess what it is by the behavior of the timing curve.

Now, the timing tables in the ECU, just like the fuel maps, are indexed by airflow and rpm. With a SAFC, this has an added effect. Since a SAFC intercepts that signal from the MAS to the ECU and modifies it, it can change the amount of airflow that the ECU "sees." If you have to correct your SAFC into the positive range, than the ECU will see more airflow Hz than the MAF is outputting, and could change the timing map you are following. The problem with this is, higher airflow levels get less timing advance for safety, and lower airflow levels get more timing advance, because the ECU thinks you are pulling in less air.

By leaning out the SAFC (big injectors, more fuel pressure, race gas) you decrease the amount of airflow that the ECU sees, and therefore you usually will get a bit more timing advance for power. This all assumes you have no knock, and also keep in mind that more timing advance will have an engine a higher propensity to knock.

I have heard of 1g guys with 660 cc/min injectors getting timing advanced as much as 28+ degrees at WOT, because you hve to pull the SAFC correction factors down a lot due to the fact that 660's are 47% bigger than the stock 450's.




That's from the tech guide. So after you went above 7 counts your timing got retarded.

I know.


i have gotten 43 counts of knock countless times and my engine still runs. great at that. i have a maft to use for injector compensation, and the afc for fine tuning. i believe my timing was pulled to 10 or 11 if i can recall, but nevertheless your right ( i think thats all you wanted to hear ) any knock exeeding 7 is bad knock. but my engine is fine is was a mistake and luckily a non catastrophic one, it is fine , maybe i will run 30 psi at the track with some 110 in her ..i will have to log it and see what is going on.

 

[compression]

He can do just fine @ 30 psi and pump gas w/ water injection. Properly delivered water injection will raise effective octane leves to around 110. That's race gas octane. I ran 30 psi on stock cams and my 18G which everyone swears is hardly bigger than an EVO3 16G. I recorded 1 degree of timing retard at torque peak.

His water injection setup is flawed:

First, he needs to run straight water as he does not have enough flow from his injector to run a water/alky mix. Running straight alky requires 3 times more flow than straight H2O.

Second, his "nozzle" or injector is not misting properly. This in turn does not give him much of the added benefit of cooling the aircharge before it gets into the combustion chamber (like what an intercooler does, but on a much grander scale).

Third, is his fuel pump flowing enough water/alky? Who knows? It's a chevy cavalier fuel pump and this is dsmtuners.com.

If he plans on running this boost often, I do suggest and intercooler upgrade. Not so much for the cooling, but for the increase in flow.

I will be running my RS60T (60-1) at 30 or so psi w/ pump gas this season, thanx to water injection.

the pump is flowing more then the injector can release hence pressurization in the lines., your right the injector isnt misting properly i will work on this , it was an old injector i found laying around is may be dirty and partly clogged this is something that needs to be looked into. i do not plan on running this boost often at all.

and straight water really? i thought a water/alky mix ( smurf piss ) is better then water no matter what the flow is?. never the less it took minimal time to make this kit and minimal cost.


<b> take a wild guess at how much it cost to make my kit</b>

 

[dsm-onster]

Heres some NACA (National Advisory Committee for Aeronautics) reports. I think this info is useful in this thread. . .

The induction of water to the inlet air as a means of internal cooling in aircraft-engine cylinders

Knock-limited performance of several internal coolants

EDIT: The above links are broken. The.pdf files are twice too big for DSMtuners upload. So go to The NASA Technical Reports Server and do a title search for the above titles.

 

[compression]

i payed 22$ for my kit. and that was for the boost activated switch. i think it works very good ecspecially for the cost and time it took to make it.

 

[compression]

retarded to 8* of timing.

you are correct it pulls 1* for every 3 counts of knock exeedin 7. but how do you know what my timing advancement is before it starts to pull my timing? it could have been well up to 28 and i had 24 knock ,it would only be retarded to 23*

24-7=17 17/3=5.66666666666667 so my timing was only being pulled 5.67*

so if my timing was even 20 then 20-5.67=14.38 so i would have had 14.38* of timing even after 24 knock. where did you get 8 from?

 

[Baron4406]

Wow I'm getting one!!! Keep in mind my mods, I have an EVO 16G, Buschur front mount, 550's, 255 walbro. I can run 20psi at 70% injector duty cycle and zero knock. However I have to either run alot of extra fuel- or pull out around 8deg of timing advance (with my PMS/Interaq) to run that. I think I'd benefit because I'd like to hit 25lbs of boost and put my timing back in. However do i need to downsize my water injection injector size because I already have a pretty effecient setup?? I was thinking of a 500 cc injector. I was gonna get this Cooling Mist setup:

http://cgi.ebay.com/ebaymotors/Wate...ewItemQQcategoryZ42604QQitemZ8051905005QQrdZ1


However this setup has more parts, a higher PSI pump, and more parts:


http://cgi.ebay.com/ebaymotors/WATE...ewItemQQcategoryZ42604QQitemZ8052436043QQrdZ1

I'll probably go with the Cooling Mist setup, however the other one is tempting too. Would a 500cc injector be alright for my setup?

 

[Slippi84]

you are correct it pulls 1* for every 3 counts of knock exeedin 7. but how do you know what my timing advancement is before it starts to pull my timing? it could have been well up to 28 and i had 24 knock ,it would only be retarded to 23*

24-7=17 17/3=5.66666666666667 so my timing was only being pulled 5.67*

so if my timing was even 20 then 20-5.67=14.38 so i would have had 14.38* of timing even after 24 knock. where did you get 8 from?

If the ecu retards for every 3 counts of knock over 7 and still includes that 7 then

24 counts of knock / 3 counts of knock per degree of timing pulled = 8 degrees of timing pulled

That's where I got 8 from

 

[compression]

Wow I'm getting one!!! Keep in mind my mods, I have an EVO 16G, Buschur front mount, 550's, 255 walbro. I can run 20psi at 70% injector duty cycle and zero knock. However I have to either run alot of extra fuel- or pull out around 8deg of timing advance (with my PMS/Interaq) to run that. I think I'd benefit because I'd like to hit 25lbs of boost and put my timing back in. However do i need to downsize my water injection injector size because I already have a pretty effecient setup?? I was thinking of a 500 cc injector. I was gonna get this Cooling Mist setup:

http://cgi.ebay.com/ebaymotors/Wate...ewItemQQcategoryZ42604QQitemZ8051905005QQrdZ1


However this setup has more parts, a higher PSI pump, and more parts:


http://cgi.ebay.com/ebaymotors/WATE...ewItemQQcategoryZ42604QQitemZ8052436043QQrdZ1

I'll probably go with the Cooling Mist setup, however the other one is tempting too. Would a 500cc injector be alright for my setup?

no.

injector size x # of cylinders x injector duty cycle = fuel flowing at that giving time


flowing fuel x 20% ( water to fuel ) = max allowed water to spray .

so you are 550x4x70%=1540cc's


1540x20%=308cc's of water allowed to inject into your engine at 70% injdc or higher.so by injecting 500cc's you would probably hydro lock your engine

 

[compression]

If the ecu retards for every 3 counts of knock over 7 and still includes that 7 then

24 counts of knock / 3 counts of knock per degree of timing pulled = 8 degrees of timing pulled

That's where I got 8 from

and 24-8= 16, so at that situation my timing would have been pulled to 16 , i misunderstood before. i though you ment it was pulled all the way down to 8* of timing