Saabscene Saab Forum - Saab Technical Information Resource banner
1 - 7 of 7 Posts

·
Registered
Joined
·
18 Posts
Discussion Starter · #1 ·
I've purchased a WI system, installed it, and am getting mediocre results. It contains a Shurflo 140-psi pump, 1-10 psi pressure switch, Aquamist 0.7 nozzle, anti-drain-back solenoid, relay, LED light for dash, and an in-dash switch for disabling the system.

The problem is that performance is no better at best and worse (lower boost and light misfiring) if the spray is activated too early. At this point I have it set for 10 psi and the performance level is about equal to no WI at all.
If I set it higher the WI won't cut in when hot, as the ECU seems to limit boost to 10 psi.

I'm getting a very good spray pattern from the Aquamist .7 nozzle, and the system is working very well. The car is a '95 9K Aero (U.S.-spec T5) with aftermarket ECU, 3" exhaust system, and K&N internal filter (stock replacement). Everything else is stock. In winter the car is a race horse, achieving >20 psi with no trouble at any time.

The problem is that the WI is not improving boost at the times I really need it, during summer months when temperatures here in Louisiana are 90°+ with high humidity. If I accelerate from low RPM (~2000) when it has just achieved operating temperature, as when the thermostat begins to open, I get decent boost, sometimes approaching 15 or more psi. However, as the heat soaks in it will usually boost only to 10 psi, then stop with no indication of reaction to detonation. At other times when detonation sets in it will boost to perhaps 13 psi, then back off to 11 or 12, presumably a reaction to detonation. It is not very consistent in this regard, but one thing is for sure--when ambient and engine temperatures are hot, boost is radically reduced. Whether or not WI is working seems irrelevant, as I can disable it at will using a swtich on the dashboard, and there is little or no difference on or off. It is only after driving for 5+ miles in very hot weather will it become limited in this fashion. I am using 93-octane (R+N/2) gas, the highest available here, and have tried different brands with no variation.

It is as if the ECU is telling it "Okay, it's too hot, stop boosting now." What sensor is instigating this? The air-temp sensor on the intake tube is about 3" downstream of the WI nozzle and I would assume the WI is affecting its reading when activated, but I think it only affects fuel injection, not boost. I am assuming the entire problem revolves around the ion-discharge knock sensor, but why doesn't the cooling effect of the WI affect this?

With the installation of the WI I was really expecting boost to climb to near-winter levels (>20 psi) even in the hottest weather, but I'm getting almost no benefit from this addition. What about adding some WSW fluid or alcohol to the mix? Any ideas would be appreciated.
 

·
Registered
Joined
·
14,305 Posts
I would start by using a smaller nozzle, Bob- try 0.5mm. The water may be affecting things adversely.

On dyno runs, my car normally shows slightly less max power (about 3%) when I'm running WI. This is because there is adequate fuelling.

However, I think what may be happening is that WI doesn't always significantly cool the intake charge. It's greatest benefit is the high latent heat of vapourisation, which leads to a significant in-cylinder temperature reduction, helping to prevent knock. This can enable higher boost pressures to be run without the APC throttling things back. The boost pressures may be the same or higher during summer, but because the density of the air is lower, you get less torque/power.

You're probably talking a minimum 20 deg C difference between summer/winter intake temps, whereas you might only see maybe 5 deg C (I'm guessing here) reduction in intake temp from the WI, so the ECU is still mapping for warm air.

In my experience if you want "normal" winter levels of boost in the summer, you need to go for a larger intercooler.
 

·
Registered
Joined
·
4,215 Posts
WI is a strange thing with the saab system
In the posts to date I do not believe anyone is reporting more hp/better boost and I think people are content to run it because of the "safety" aspect of it
however you have maybe stumbled on the answer..
the water could indeed be affecting the ion sensor
mechanism of the trionic boost control..

I have noticed that my car always goes better when it is raining probably due to the cooling effect of the increased relative humidity knocking back the knock...
In water injection the water is injected in fine droplets just before the inlet manifold typically.
these droplets may not have time to fully atomise or even break up and so have little effect on the knock in the cylinder or even these discrete drops of water droplets form conduction paths in the charge so affecting the ion sensing arrangement of the trionic system

Maybe a better place would be pre-intercooler where the charge temperature is higher and could volatilise the water so cooling the charge ..
and it would be fully atomised before it is drawn (pushed) into the cylinder ..

I would suggest its worth a try ...
 

·
Registered
Joined
·
1,448 Posts
Try posting your questions here:

http://www.aquamist.co.uk/phpBB2/index.php

Aquamist was kind enough to host the old TurboICE forum. Soooo much good information there.

To post an answer in brief...

Firstly, water injection MUST be tuned. If you are running very low octane (like I am here in Southern California) the engine will partially tune itself to the water injection. If you are already running sufficient octane, or are using an "off-the-shelf" aftermarket ECU, you may not experience much advantage without tuning.

In order to tune for water injection you must do two things:

1. Advance the timing. Many tuners are afraid to do this, partly because they aren't used to advancing the timing so much, and partly because they worry about what happens if the WI fails. Your ECU should be able to handle a failure though, as it would be no worse than running into a lower octane fuel by accident. It should be just fine.

The reason for the increased timing is due to the "Flame Development Angle". This does not affect overal burn rate or power because it does not change the "Rapid Burn Angle". The Rapid Burn Angle is how fast the charge burns, whereas the Flame Development Angle is how long it takes between when the spark is fired, and when the flame wave forms and moves away.

2. Lean the mixture. This is something else tuners often refuse to do. You will probably frequently hear them say "well water doesn't burn" ... well neither does any fuel beyond a 14:1 Air/Fuel ratio. Most people tune for about 13:1 ... because of the affect of water, on a Wideband O2 sensor that will appear as 14:1. When using water injection O2 sensors may read slightly leaner than they actually are. This is why you should not use WI when cruising as it could affect emmissions and mileage.

Leaning the mixture is actually a desirable thing to do if you can get away with it. It brings up the Flame Development Angle and increases the Rapid Burn Angle (which increases power). Also a leaner mixture generally results in a much cleaner burn as water does not pollute.

Also nearly all WWII aircraft used water injection. It was termed "anti-detonation injection" and was basically a 50/50 mixture of water and methanol.

At any rate, check out that forum. There is more information there on Water Injection than anywhere else I know of.

Enjoy!

Dubbya~
 

·
Registered
Joined
·
18 Posts
Discussion Starter · #5 ·
To Adrian: I wasn't under the impression the T5 ignition timing could be altered without a remap of the ECU.

Update: A friend emailed me and suggested that I check my plug gap, which I found at .042", on the high side of the spec. I backed it down to .039" (about 1mm) and it made a big difference. I also added some ready-mix WSW fluid (mostly water) with methanol. Not sure which change made the bigger difference, but a test run revealed no more 10-psi brick wall as before when hot. Rather, I was getting 13-17 psi, then detonation and a back-off of a few PSI, then boost would escalate again as RPM's increased (no more than 4000 during this test). This is encouraging.

In the future I will be testing with iso alcohol and methanol for the best mix. I'll also check out the Aquamist link.
 

·
Registered
Joined
·
1,448 Posts
To Adrian: I wasn't under the impression the T5 ignition timing could be altered without a remap of the ECU.

That's the idea. I was just trying to give you some direction in case you need to interact with a specific tuner about water injection.

If the engine is sensing enough detonation to back off the boost adding water injection may raise the knock limit enough that the engine advances timing, leans the mixture and raises boost. Which is exactly what you want! Unfortunately it's hard to get it to work the way you want it to without a remap.

Dubbya~
 

·
Registered
Joined
·
1,044 Posts
In the future I will be testing with iso alcohol  [/b]
be careful if its a genuine Aquamist pump as they clearly state in the instructions not to use any other type of alcohol except methanol.....the seals aren't compatable and the pump will seize.
 
1 - 7 of 7 Posts
Top