1GGTE water injection (IAT sensor question)

[amon]

G,day all, i am about to install some WI to my 1g.

Question 1. My IAT sensor i believe is located at the mouth of my AFM,does this sensor have to be downstream from WI nozzle to read the effect from the WI.

Can/should i move it(some how) into that location as i dont particulary want to go pre AFM?
Could the AFM cope with a bit of a wash?
If this is not possible is there a sensor(A/market)that i can use with my standard ECU.

All the reading i have done leads me to believe that the sensor needs to be after the nozzle,any thoughts?

The reason to go WI was to rid myself of myriad of piping and IC and due to extreme humidity and very high temps in darwin, try to lower intake charge temp lower than i could with just the cooler.
The difference to performace the car is considerable from Wet season(hot)to the dry.

Thanks in advance

Guy

[timbosaurus]

Is there a temp sensor in the afm? Odd they would put one in pre Turbo, but would explain why there are so many wires in it. Never really looked into it, as I plan on ditching it asap.

IAT is used (I presume) to calculate oxygen content of the incoming volume of air. Thinking out loud, maybe IAT needs to be at the point where air volume is measured. In the case of afm, it is measured at the afm, but when u use map sensor u measure IAT post Turbo.

I could be completely on the wrong track, but maybe a thought starter...

[Kedderz]

Stock computer I assume as your using afm?

[amon]

Yes,but i would love to rid myself of that too. one thing at a time

[Duk]

From what I've read about water injection over the years is that you DON'T want the inlet air temp sensor to see the post water injection air because water can/does give a false cooling effect to the sensor itself and so a false reading.

Inlet air temp sensors inside the AFM are more to allow for the changes in ambient air temperatures rather than to measure the actual air temperature going into the engine post intercooler. From a manufacturer's point of view it kinda makes sense, the intercoolers performance is basically consistent with ambient air temperature in regular road driving + a big fat safety margin of over fueling and less than ideal ignition timing.

[Supra967]

what are you doing with the reading? Is it just for logging temp/hooking up to a guage? I'd just put another sensor downstream. Are there map sensors with inbuilt temp sensors? You could hook up a map/temp sensor in the intake and log boost and temp, then it's already set up if you go standalone in the future

[amon]

Hey fellas,
Duk;thats interesting as i have been trawling the net and have been reading the opposite,meaning that the sensor would measure the new intake charge temp and would adjust the the fuel and maybe some timing.
i dont know how else the ECU would sense it was a say,a real cold day
Or that it does not matter to the ECU and it just gets a bigger bang for same fuel and a cooler denser charge

Supra967:i thought that the sensor that was in my afm told the ECU how cold the inlet temps and adjusted fuel there are no other temp sensors besides coolant
Thanks lads

[Duk]

My understanding is that the wet sensor can read a lower than an actually correct temperature.
The other problem is also the fact that while the inlet air is cooler it doesn't have much if anymore (air) density because the air is now more humid.

While I never tried it with my SC'd AW11, I had thought of having at least the factory IAT sensor (I also thought about having the whole AFM) after the intercooler and water injection. The logic being that the IAT sensor would measure the true inlet air temperatures that allowed better for heat soak of the intercooler.
The reason why I never tried it was simple, even if I could get post intercooler air down to ambient temperatures with the water injection, there would still be no real benefit. And if I couldn't get the inlet air down to ambient, then the IAT sensor would be telling the computer that the weather was hotter than it really was and the computer would deliver even less ignition advance.

Having said all of that, the best effect of water injection is in fact that it acts more like an octane booster (detonation suppressant) because of the amount of excess heat it absorbs during the combustion process rather than the heat it absorbs by just being sprayed inside the intercooler plumbing and/or plenum chamber.
Ideally a good water injection equipped engine should have more ignition advance based on the fact that water injection is operating rather than just measuring the IAT cooling effect it has. Basically the ignition map would be done in the same way it would be done for a higher octane fuel.
Those in the know (like the old NACA reports) always talk of water to fuel ratios, rather than say, 'inlet air temperatures to water ratios' or 'air flow rates to water ratios'.

*** A slightly relevant update. When I 1st went with my twincharger installation, I had the bigger than standard AFM pre turbo like most people would. Unfortunately that had quite a lot of bends in the pre turbo plumbing and 1 was so tight I had to make a lobster back bend to get the job done. My water injection nozzle stayed after the throttle body but before the supercharger. When I made my new tubular exhaust manifold and collector style inlet manifold I then fitted the AFM in the plumbing between the turbo and the throttle body (blow through style) but I removed the inlet air temperature sensor and placed it in the turbo's air inlet so that it wasn't measuring the heated air from the turbo. I used a better water spraying nozzle after the intercooler. I never had any detonation with either set up except for when I didn't seal the Pulsar GTiR intercooler to the engine lid and air could go around the 'cooler instead of through it (but I wasn't using WI at that time), and while the car has never seen a dyno, it would swallow more ignition advance (I used a Trust E-Manage Ultimate) even when 18-odd psi of boost was being shoved down it throat with 12.5:1 full load AFR's.***

[thechuckster]

Inlet air temp sensors inside the AFM are more to allow for the changes in ambient air temperatures rather than to measure the actual air temperature going into the engine post intercooler.

no, the AFM temp sensor only measures the temp of the air entering the engine as the only way in for air is via the AFM. Based on the RPM, air-temp and calculated mass of air entering the engine, the ECU calculates the fuel required for that mass of air entering the engine.

From a manufacturer's point of view it kinda makes sense, the intercoolers performance is basically consistent with ambient air temperature in regular road driving + a big fat safety margin of over fueling and less than ideal ignition timing.

sort of... it knows how much air has entered the induction system (again, based on the AIT and what the AFM vane has been doing) but not how much cooling has occurred in the intercooler, so it takes an educated guess at the current performance of the intercooler (based on current air-flow rates and engine speed).

Cooler air is denser and thus contains more oxygen - with which it can then burn more fuel and run higher ignition advance to extract more energy from the combustion event - thus making more power.

If you want the ECU to take advantage of the cooler/denser air-charge entering the engine, you have to measure it.

Ergo, if using water-injection you should measure the air-temp downstream of the water injectors (and far away from the injectors) and ideally in the plenum.

[Duk]

no, the AFM temp sensor only measures the temp of the air entering (ambient air) the engine as the only way in for air is via the AFM. Based on the RPM, air-temp and calculated mass of air entering the engine, the ECU calculates the fuel required for that mass of air entering the engine.

Ah yeah, isn't that what I said?

sort of... it knows how much air has entered the induction system (again, based on the AIT and what the AFM vane has been doing) but not how much cooling has occurred in the intercooler, so it takes an educated guess at the current performance of the intercooler (based on current air-flow rates and engine speed).

I think you may be giving some these older engine management systems waaay to much credit. Given what little processing power they had, how do you expect them to be able to take any sort of guess as to the state of heat soak the intercooler me be suffering from? Surely if they had that spare capacity, adding another sensor to measure post intercooler temperatures would be better suited to real world conditions.

Cooler air is denser and thus contains more oxygen - with which it can then burn more fuel and run higher ignition advance to extract more energy from the combustion event - thus making more power.

While the air may well be cooler, its density hasn't gone up by much (no additional air molecules) because of all of the newly added moisture. Cool dry air is better for performance than cool damp air.

[thechuckster]

Ah yeah, isn't that what I said?

sorry - i thought you were referring to air temps in general - air outside the car will be different to what's in the engine bay and even what might be going into the AFM etc.

I think you may be giving some these older engine management systems waaay to much credit. Given what little processing power they had, how do you expect them to be able to take any sort of guess as to the state of heat soak the intercooler me be suffering from? Surely if they had that spare capacity, adding another sensor to measure post intercooler temperatures would be better suited to real world conditions.

I'd expect them to have a basic compensation table that uses rpm and afm to return an arbitrary cooling performance value for the intercooler. Assuming a fixed temp drop across an intercooler would have been a recipe for disaster.

Toyota was a fan of AFM or mass air flow metering systems (which makes it easier in some ways to adapt a proven EFI system to a lot of engines). This system design made any post AFM temp sensing pointless as a primary driver of fuel calculation is the volume of air entering the engine.

Once you get to a MAP sensored engine, the post-intercooler temps are very important in determining air density.

While the air may well be cooler, its density hasn't gone up by much (no additional air molecules) because of all of the newly added moisture. Cool dry air is better for performance than cool damp air.

My grasp of physics is less than ideal these days so bare with me: by cooling the air you will get more oxygen molecules in a given volume, the H2O molecules don't replace the space 'given up' by the oxygen. Yes, in a given volume of air you will now have more H2O than before. As the system is closed between AFM and combustion chamber, you don't get extra oxygen but you have a cooler and denser mass of air - and at a lower pressure than if you didn't inject water.

My own experience of a water-injection was that I could shove a lot more boost in the engine and when you made the injection liquid 1:1 of water & methanol, it was even more damn impressive.

Also I think justen's GT8 proves that cool air and water injection equates to tyre-shredding performance.

[Supra967]

I think what Duk is getting at (assuming post charger WI) is that the same amount of air has entered and exited the charger. Cooling the air will not create 'more' air - conservation of mass. So for a fixed speed SC the boost pressure should go down post WI. Turbo would be different depending on waste gate reference point. So goes the theory but it'd be interesting to any actual manifold density changes in the air, given it takes time for the water to absorb heat from the air. I'm with Duk on the cylinder and combustion chamber heat absorbtion.

Regardless, the 1G won't be able to sense the WI changes. You could guess, much like the factory comp, at post AFM conditions and adjust your base timing a little? You should have a little safety net if you're using 98ron fuel anyway.

[Duk]

1 of the best books I've read on the subject of water injection was simply called 'Supercharging!' written by Greg Gordon. While it is primarily aimed at supercharging the old Alfa Romeo V6 engines, the section on water injection is quite extensive and is based on a lot of data that Greg got from old (finally released to the public) NACA documentation of research done for fighter and bomber aircraft during world war 2. The thermodynamics of water injection was very well studied by NACA in those days.
Now it is very easy to misinterpret and mis-quote stuff, but I will quote a few things here:

"The great majority of the cooling from water injection takes place during the compression stroke when the valves are closed. Very little takes place prior to the intake valve. The density gains in what cooling takes place outside of the cylinder are partially offset by the increase in humidity.

Of course NACA did a lot of tests with water injection. They found that the density of the air only slightly increased with water injection. Here is what they said, "The fact that the amount of air mass inducted increased so slightly with the quantity of water injected seems to indicate that the vaporization was taking place within the the cylinder rather than within the inlet pipe."

Injecting of water into the engine's intake system has an effect that's very analogous to using higher-octane fuel. Just how much of an effect does it have? According to NACA, a lot. They came up with the following numbers. An engine normally requiring 100 octane fuel can operate satisfactorily on 80 octane fuel with a water/fuel ratio of 0.6:1. The same engine could operate on 88 octane with a 0.4:1 water/fuel ratio or 94 octane with a 0.2:1 ratio."

So while am definitely encouraging the use of water injection (hell, I've used it and will use it again on my twin charged AW11 and eventually on my Vortech fed Alfa Romeo 3 litre 75), I think that more accurate interpretation of what is going on needs to be done to help all users of water injection.
For my Adaptronic equipped AW11, I plan to have a water injection fuel (full load at 12.5:1) and ignition maps that get the best of what I can on 98 octane fuel with a decent safety margin. Rate of water injection will be based on injector duty cycle used as a signal to control an FJO water injection solenoid.
For the non-water injection maps I'll run richer and with safer ignition timing.
The inlet air temperatur sensor will be behind the point of water injection so as to only see the temperature of the air coming out of the poor, struggling Pulsar GTiR intercooler (with a fan underneath it ).

[thechuckster]

what about injecting the water pre-intercooler? or is there a significant lag time between start of injection and arrival of H2O in the inlet manifold if done that way?

fwiw: my old setup was injecting the water/meth mix pre-turbo (made for very simple plumbing)

[Duk]

I was injecting my water pre supercharger for quite a while. There are plenty of people, including Greg, who suggest to not pass the water through an intercooler. I did it to try and help the supercharger seal and because the AW11 intercooler is in such a poor spot, I figured it didn't matter.

The other thing to consider is that with an increase in pressure, the boiling point of water increases. How that effects the waters ability to vaporize and evaporate I don't know.
There was a good Youtube video of a Cosworth Sierra that had a custom made plenum chamber with a clear front end plate. The car was being run (on the dyno I'm guessing) and you could see masses of water collecting on the clear end plate. I recon the car was featured in Zoom many years ago and if it was the same car, the guy was using an Aquamist system.