Major differences between in-port and post IC injection?

[max_torque]

I have been trying to understand the main differences in running with direct in-port (ie injecting in intake runner of each port) and post intercooler injection in conjunction with a vehicles std EMS.

I have made 1 major assumption, that at the point of ignition all the water has been vapourised due to the high temp and pressure in-cylinder for both injection positions(is this valid??)

If the air mass flow, AFR, and WFR, turbo / IC efficiency & effectiveness are the same for each method, then the charge temperature at the point of ignition will be similar? (need to further understand the heat transfer to combustion charge along the length of the intake system / chamber, ie does in port injection actually cool the head metal temps etc?)

However, for the post IC injection, the engines' management system will be able to "detect" the presence of water via the reduction in plenum air intake temperature, so adding fuel and advancing ignition timing. This will take advantage of the cooler charge, and combined with a higher manifold Vol eff due to the denser plenum charge produce more torque.

With direct to port injection, the EMS cannot sense this, so there is unlikely to be any power gain on a std EMS system, unless fitted with 2 things.
1) active and aggressive knock control (ie can advance spark agressively till knock is sensed (typically this is now adaptive (ie Bosch ME*.* etc), and may take some time to learn the new knock threshold)
2) EGT sensing, Again some applications of Bosch / siemens ems run with pre turbine EGT sensors to control component protection overfuelling.

Does this explain why people typically see a greater benifit from post IC water injection on road cars without specifically mapped EMS?

[Richard L]

Direct port injection is good for inlet valve cooling and precise metering. Most direct injection system uses a narrow angle jet pointed at the inlet valves. It is to maximising the cooling effect of water at those selected components.

As mentioned, without the use of some engine management mapped or a more intelligent ecu, the engine will not be able to take advantage of the effect of water. It will all depend on what the car is used for and depend on how far you want to go. Recently more aftermarket tuning tools are available but only a few willing to tune WI with it

WI on a standard road car should have post or pre IC injection, most modern management (un-flashed) will be able to learn new knock threshold at trim the timing to suit. Unfortunately WI injection seemed to be placed at the end of the list of performance modifications, by the time their turn has arrived, the engine managment has already been blocked out to do anything useful.

A standard road car straight out of the factory has certain amount of safeguard inbuilt such as EGT, knock retard and lambda sensor. Unfortunately some safeguard is being overdone - fuel dumping is the favourite because it is readily available and more effective than retarding ignition and dropping boost. If water inejction is avaiable as a standard fitment to combat those adverse conditions, it will be the most ideal tool to produce a "low emission" car with plenty of power.

Read this link (http://www.aquamist.co.uk/dc/coollinks3/index/rally/saab/press/press.html) when SAAB published the results of WI on a eco engine. I have been closely following the project right from the start. It used port injection and produced great results. Pity it never got beyond the gates of the R&D site in Sweden.

For the entire bundle to work properly, the engine managment must be fully involved. At present only a few tuners are able to tune with aftermarklet ECUs with water injection effectively and the other 99% just repeating what the car makers were doing for many years - plus a few more pounds of boost with richer a/f ratio- not very encouraging and innovative.

One day the accountant in these big car makers will bow to the pressure of the environmental group and the performance enthusiasts to implementing water injection as standard - it just make good sense.

My answer to your question would be yes, if the orignal ECU has not been hacked and modified - for a standard production car.

[max_torque]

A standard road car straight out of the factory has certain amount of safeguard inbuilt such as EGT, knock retard and lambda sensor. Unfortunately some safeguard is being overdone

Yeah, i know quite a lot about this being an automotive calibration engineer for the last 10 years!

On Focus RS for instance the engine is mapped to BLD -3 deg and LBT - 3 AFR (runs at richer than 10:1 in catalyst protection mode!)

Water injection is only capable of small improvements in emissions and fuel economy over the mandated drive cycles as these are at relatively low load for a passenger car. (WI in conjunction with a downsized, high compression turbo engine has some merit though)

"real world" fuel economy however in the cut and thrust (especially in europe with our point and squirt driving) could be significantly improved by WI limiting component protection overfuelling.

[Richard L]

Water injection is only capable of small improvements in emissions and fuel economy over the mandated drive cycles as these are at relatively low load for a passenger car. (WI in conjunction with a downsized, high compression turbo engine has some merit though)

"real world" fuel economy however in the cut and thrust (especially in europe with our point and squirt driving) could be significantly improved by WI limiting component protection overfuelling.

I entirely agree WI will only improve the fuel and emmission on near WOT applications for most passenger cars.

The only way to improve the whole range is to run the gasoline engine on an excess-air mode - similar to the diesel. It will require an electronic throttle that switches to WOT during cruising and just add enough fuel to maintain lambda at 1. the In-effciciency of the intake tract is almost eliminated. Water is then injected to keep the in-cylinder temperature to a sensible level - the quanity injected will be just to keep the air moist - moist air will lessen the octane requirement of the engine.

[max_torque]

The basic problem with gasoline engines is not there emisisons based on polutants such as CO, Nox, or THc, it's there fuel economy and CO2 emissions. An EU4 car (reqired by european law from next year) must emit less than 1g/km of CO, 0.1g/km of THc, and 0.08 g/km of NOx. It will also be running at lambda 1 within approx 11 secs of start (at the 25 degC test temp) whichit will stay at for the whole test duration unless in "overrun" when the fuelling will be totally switched off.

The basic problem is that the current technology of 3 way catalytic converts have to be within approx 0.05% of lambda 1 to efficiently convert the pollutants, this means we cannot run lean of lambda 1 for fuel economy. The next major step will (assuming there is not a major on-cost reduction of NOx absorbers within the next few years) therefore be downsizing, ie if you have to be at lambda 1, then you need to have the smallest airflow possible to limit the fuel flow. But the market demands a reasonable performance, say 100 to 150 bhp per tonne typically, and 150Nm per tonne. this can only be obtained through forced induction, which requires a low geometric CR and massive overfuelling!

[lokmeup]

I am running a Aquamist/Shurflow direct port WI system on my 2g Talon with extremely good results. My web page gives a quick description of the way mine is set up http://lokmeup.iwolfe.com/Talon/WaterInjection.htm

I am running 26-28psi, at a WBO2 reading of 11.4-1AFR, corrected 11.8:1AFR, at +3* timing, with zero to little knock retard (~.4), flowing 58-59lbs/min airflow throught the engine and this is all on 91octane pumpgas.

Before going direct port, I was running the same setup with approximately the same amount of H2O/Meth injection, but was getting 3* of knock retard when I tried these higher boost pressures.

..for me, direct port made ALL THE DIFFERENCE!!

Ron