Tuning for water injection: fuel, ignition, and EGT

[Richard L]

Hotrod has posted this link on another forum, it is extremely good reading... recommended

http://enginehistory.org/Frank%20WalkerWeb1.pdf

Richard

[JohnA]

Good reading

I guess some of Frank's last inventions (end pages) might be of less interest to this forum:
A non-slip stacked paper fasteners
A urinary tract catheter that allows flushing and treatment of the prostate gland. :lol:

[Richard L]

I was very surprised about the isopropyl being useless. :shock:

[JohnA]

Originally Posted by Richard L

I was very surprised about the isopropyl being useless. :shock:

In their setup, it was useless because it was making the mixture too rich to ignite.

This may not be the case with our setups (injection, ECUs, widebands, etc) :wink:

[simple]

I had been reading thru this topic few times now trying to get a grip. I have some thoughts to share.
First, talking about tuning new car for WI. It has been mentioned that you want AFR about 12.5 as cars under boost run normally richer. Well here is a third gear pull of my car. It?s totally stock:



AUDI 2002 model 1.8 Turbo
Max boost hit early about 7PSI@2000RPM. After 4500RPM there is boost reduction due to turbocharger going outside efficiency range at the same time AFR seem to be becoming richer. This does not correspond with max torque band as it is 2500-3500RPM.
Look like new cars run lean stock. My cruising AFR is 14.7@ light load(80Km/h)

Second, according to stock ECU knock sensor reaction KNOCK been detected throughout whole RPM range. Under knock I mean not audible knocking but early knock detection by ECU(mild knock). Here is Correction Factor data log:



CF1,2,3,4- timing pull back by ECU due to high knock sensor activity.1-cylinder number one and so on.
BTDC ? ignition advance.
You can see as CF goes up ECU pulls ignition down. This present equally thru whore RPM range not just thru max power band. This would mean that WI have to be active thru whore RPM range. During normal cruise there is no CF noted. It only appears under the boost and can vary from during whole RPM range to showing up here and there. There is no consistency in knock. Looks like it is very unpredictable but load dependand. It is also noted that poor fuel quality will make it more intence as well

Any comments anyone?

[JohnA]

Originally Posted by simple

...First, talking about tuning new car for WI. It has been mentioned that you want AFR about 12.5 as cars under boost run normally richer. Well here is a third gear pull of my car. It?s totally stock:

We are not talking stock cars here mate, the manufacturer has optimised all parameters during stock operation.
If you inject water on a stock car you are unlikely to make any more power. Most likely you will *lose* some.

After 4500RPM there is boost reduction due to turbocharger going outside efficiency range....

How do you know that? Are you datalogging the solenoid's duty rate?

Look like new cars run lean stock. My cruising AFR is 14.7@ light load(80Km/h)

That's how it's meant to be.
It is high throttle/high load conditions that AFRs are meant to go richer.

Second, according to stock ECU knock sensor reaction KNOCK been detected throughout whole RPM range. Under knock I mean not audible knocking but early knock detection by ECU(mild knock).

Are you talking about the knock sensor output, or the ECU's interpretation of Knock? These are two different things, the sensor produces output all the time, the ECU has filtering algorythms to make sense of it all.

...It is also noted that poor fuel quality will make it more intence as well

this is very true.
However, I doubt that you are getting any knock on a stock car running stock boost and the recommended fuel. Even if you do it will be on a very hot day and at max torque revs if you boot it real hard. The ECU will adjust timing a bit and that's that.

[simple]

Originally Posted by JohnA

We are not talking stock cars here mate, the manufacturer has optimised all parameters during stock operation.
If you inject water on a stock car you are unlikely to make any more power. Most likely you will *lose* some.

Understood. This means that levels leaner that 12.5 will be neded. I refer to this part to indicate that 12.5 can be not "lean enogh" for some WI applications.

How do you know that? Are you datalogging the solenoid's duty rate?

I datalog solenoid as well. It is ECU who pulls boost down by changing duty cycle. AUDI uses small KO3 for 1.8 engine. That how I manage to get 7PSI @ 2000RPM. According to AUDI tuners, KO3 going outside it's effcency at 5000RPM so boost is reduced to keep it with in the map's parameters. I can do a plot for you on the KO3 map if you are interested

That's how it's meant to be.
It is high throttle/high load conditions that AFRs are meant to go richer.

Understood. I mentioned it here as piople where stating 14.5 as lean "low load" operation

Are you talking about the knock sensor output, or the ECU's interpretation of Knock? These are two different things, the sensor produces output all the time, the ECU has filtering algorythms to make sense of it all.

Trying to extract algorythms is what we are working on now. This graph contains V (voltage) of each knock sensor for each cylinder. I was refereng to KNOCK as fackt not the bakground noise.

this is very true.
However, I doubt that you are getting any knock on a stock car running stock boost and the recommended fuel. Even if you do it will be on a very hot day and at max torque revs if you boot it real hard. The ECU will adjust timing a bit and that's that.

AUDI has ECU which is slow to adapt. As I always use high RON fuel it adopted to it. Accaisinally I put "normal" fuel recomended by AUDi for datalog purpuse and i notice that a lot more knock is present on datalog. It takes about one full tank of fuel for ECU to re-adapt to new petrol and ajust it's map to suite.

[JohnA]

Originally Posted by simple

Understood. This means that levels leaner that 12.5 will be neded. I refer to this part to indicate that 12.5 can be not "lean enogh" for some WI applications.

You need to run more *boost* as well.
If you stay within stock levels then power gains will be minimal.

...I datalog solenoid as well. It is ECU who pulls boost down by changing duty cycle. AUDI uses small KO3 for 1.8 engine. That how I manage to get 7PSI @ 2000RPM. According to AUDI tuners, KO3 going outside it's effcency at 5000RPM so boost is reduced to keep it with in the map's parameters.

One way to push the turbo beyond the 'normal' limits, without compromising efficiency is to inject before the turbo, at the 'eye' of the compressor if possible. It is regarded as experimental, but boy does it work.

...AUDI has ECU which is slow to adapt. As I always use high RON fuel it adopted to it. Accaisinally I put "normal" fuel recomended by AUDi for datalog purpuse and i notice that a lot more knock is present on datalog. It takes about one full tank of fuel for ECU to re-adapt to new petrol and ajust it's map to suite.

That is slow indeed. They tend to be conservative, protecting the engine if fuel is suspect, we can't blame them for that, can we? :wink:

Try injecting 50/50 water/methanol and see how it goes. Even with a small jet (0.4mm) operating at 4psi (pretty low boost level by the standards here) knock should disappear, even if you push boost up to 1 bar.
If you are still seeing knock, then it's probably noise that you're registering, not real knock.
...or your fuel is of very poor quality indeed :shock:

[Boostingawd]

Originally Posted by b_boy

Now one thing I've notice in the various "calculators" I've browsed is that more power can be had with lower AFR than 12.5. While this is the fuel ratio of optimal power per gram of fuel, it may not in practice be the AFR of best power. So, I question the assertion that 12.5 AFR should necessarily be the starting goal for best power.

I would think that a safe AFR would be based on the amount of meth injected. Not to sure how a mixture of water and meth would play into effect since I have only run meth. Stoich for gasoline and meth are different. With that in mind more meth means a richer AFR. Well at least this is what I see on my car, the only problem I have is that I am spraying the same amount of meth at all boost levels, since it is a single stage injection that turns on at 20psi. It does seem to have a bit of lag down low and maybe it would be smart to run a leaner mixture at lower RPMs to help with spoolup. It is something I have been thinking about but ran out of time last year since my car is not driven in the winter.

When it get nice out I am going to attempt to run one more jet and hopefully 3-4 more pounds of boost.

Victor

97 GSX, 32 psi with 1.0 & 0.8mm jets @ 100psi with 93 octane, roughly 12.8GPH.

[Airspeed]

Hello,

First of all, this is a great post!

For me, this graph below has been very informative with regard where to tune for.



regards,
Walter