EvoTio

iTrader: (10)

On my Evo installation, I connected the Red wire with the 3 amp fuse to a 12 volt unswitched source under my dash (the wire loom to my HKS Turbo Timer). The other red wire was hooked up to the HSV valve. The grey wire does not see a direct 12 volt source, you hook that up to the 806-162 water pressure switch. I've found out that the 806-157 pressure switch trigured my water injection to come on at about 11 PSI. I turned it down to come on at about 10 PSI which my EVO seems to like especially in the part throttle range. I also hooked my Green wire from the FiA2 to the #1 injector wire from the ECU to obtain a injector source. That worked out real well. I've run a 50/50 methanol/water mix on my Talon for a couple of years prior to installing the 2D on my Evo with no problems. I hope this info can help you out.

Richard L

iTrader: (5)

Our technician have added an extra red wire (same) to assist installation bur forgot to update the wiring diagram, here is the updated 2d drawing. The two red wires are interchangable.

You must not connect the grey wire to the +12V, it should go to the 806-162 pressure switch (grey wired terminal). The system is activated when the 806-157 manifold pressure switch goes open circuit.

EvoTio is 100% correct on his reply.

Richard L

iTrader: (5)

Thanks for helping out

SlowCar

iTrader: (4)

I am putting together a hybrid shurflo/aquamist water injection system so I would not have the 806-162 pressure switch. Where would the grey wire be connected to?

The grey wire looks like a trigger wire....for arming the FiA2 (when preset water pressure has been reached in 806-162; from a N/O to a closed circuit) Thus giving the FiA2 a +12V signal?

Richard L

iTrader: (5)

Hybrid the Shurflo pump to an aquamist 2d upgrade system is a nice way to inject water at rate closely related to your injector's duty cycle.

The grey wire is a problem. Here is the reason why. The Grey wire from the FIA2 is for detecting the duty cycle of the aquamist pump, the duty cycle signal of the fuel injector is already reside inside the FIA2. The FIA2's "water fault" diagnostic circuitry looks at these two signals in real time and compares them, if one of the signal is out of sync with the other, it will flag out a "fault" on the orange wire (switch to ground).

The Shurflo pump works in a different way to the aquamist pump, it doesn't have a signal that is related to the delivery rate. Without such signal, the FIA2 (grey wire) cannot made a real time comparison so it will throw out a fault.

Do you run you shurflo pump continuously? Tell nme as much as you can and see if I can find a way around this problem.

You can invest on a flow sensing system (DDS3) to monitor the flow and control your boost that way.

Richard L

iTrader: (5)

Thirteen pages - a lot to absorb. Thanks for the link.

My first impression on the system is good, every car should have such a system.
Let me try to simplify on what I have read and understood.

The Xede is basically a sensor signal manipulator, resides between the sensors and the factory ECU. Two main signals are being intercepted - load and crank angle sensor (for detecting RPM and igniton timing modification). It also have an 0-5V analogue input where one can use it to to create extra tables for fuel or ignition trim.

It also has a few outputs to switch relays etc and also have one PWM output sync to engine speed. The relay output can switch on and off a water spray or nitrous etc. The PWM output is intended to use to drive a 5th injector (I think) but not quite suitable to drive the Aquamist's HSV due to lack of resolution at high engine speed. There is also a PWM boost valve output to enable user to alter's the factory boost level.

Please correct if I got it totally wrong on Xede specifications above.

The S.M.A.R.T. idea is sound. I can't help thinking why it is repeating something the engine' brain is already doing, it processes the knock signal and makes timing adjustment to suit the knock thresholds of the fuel in your tank. Fuel trimming is also something already reside in the ECU, under the extreme operating conditions, the ECU dumps fuel to protect the engine.

You can argue that the Xede can re-scale fuel injetors but being a front end system, it merely trims the load signal from the MAP sensor or MAF sensor, it is not the same as a reflash. Utilising a simple knock sensor amplifier to complete the SMART trim is a good idea, impovement on that department is a must because isolation the true knock signal takes more than a simple circuit - it is a very involved subject and some car manfacturers uses multiple "tuned" knock sensor and DSP chip to retrieve the true knock signal. Without such intense processor power, I cannot see that the SMART system can untilise this signal to trim fuel and ignition consistantly. Over-active knock signal will not be good for power.

I still believe the way forward is to use the Xede as intended, it is an excellent interceptor. The SMART system will be good if Vishnu can spend a bit more time on the knock signal processing circuitry. It will be a challenge to exceed the factory ecu's "knock processing"capability.

I failed to find a "back to back" dyno plot with the system switched "on and off" on that thread. I am open minded.

SlowCar

iTrader: (4)

The shurflo pump has a 100psi pressure switch and i will be attaching an accumulator to it. The pump will run very litttle as the accumulator has a capacity of about 700cc.

xiaoni420

iTrader: (32)

Abner!!!!!!!!!!!!!!!!!!!!

honki24

iTrader: (23)

I run a custom WI setup (built from coolingmist.com parts) w/ 15% Water/fuel, triggered by boost pressure. I have it set to start @ 4psi because of the lag between when the sensor feels it to when the pump actually gets the water into my intake pipe. In that 1-1.5 seconds I'm already hitting 15-20 psi. I haven't gone to the dyno yet, but I will sometime soon, plan on running a couple times w/ in on and a couple w/ it off.

I have a question for all you brilliant people though... Buschur Racing has made a "chip" (likely just a resistor or some other simple mechanism) that keeps the engine temp signal within the 173-190 degree mark. Anyone know how I can linearize the temp-voltage correlation on the EVO? And even furthur, anyone know what simple electronic element(s) I would need in order to keep the upper limit of voltage below the corresponding 190 degree mark?

>>Reason those questions are relevant: My car has a dynoflash, which I know is around 11.4 AFR. W/ my WI I know that I need to go a little leaner to make more power. This is why I want to negate the coolant temp enriching that is induced by the ecu when the car reads over 190ish.<<

Richard L

iTrader: (5)

At such slow cycling rate, the aquamist 2d controller will see it as a water fault since the pump is being idle for so long. What if you reduce the size of your accumulator to 30cc? if the cycling rate is faster, it might work.

Richard

Richard L

iTrader: (5)

Can you keep the coolmist pump running contiuously or using a 700cc accumulator (as slowcar) and use your pressure swich to energise a solenoid valve - it will be much quicker.

The engine coolant sensor reads 2-3K ohm when engine is cold (32-50F) and drops to 200-400 ohm when reaching proper operating temperature (190F ish).

This translates in volatge of 3-5V (cold) and 0.5-1V when wramed up.

Are you trying to fool the ECU to cut fuel by droping the voltage further to, say 0.3v to 0.8v? I am not sure if the is will work as the sensor (thermistor) is not linear - I would like to guess what algorithm the ECU used to linearise the sensor.

Richard

honki24

iTrader: (23)

firstly, no I cannot keep the pump running continuously, and I do not use an accumulator, the expansion in the tubes provides sufficient 'lagging spray' as it were. (under pressure they expand, when the pump cycles off, they contract and water flows out @ nearly the same rate)
EDIT: now I understand what you are saying, yes, that does sound like a much 'better' idea, but my aim was for the system to be cost conservative. I won't quote the price, b/c aquamist's cust. service is excelent and I don't want to take away from them, but I built and installed the whole thing for very little money.


Secondly, thank you very much for the info, are you absolutely sure that it is correct? What I'm saying is that I'm taking your info (200-400 ohms/ .5-1 Volts when warm) as rock solid info on which to build. if possible, I'd like to know the exact values for the corresponding temperature of 280 degrees farenheit. I am not trying to trick the ecu into pulling fuel, but rather trying to stop it from adding fuel when coolant temps rise too high. (b/c the water will cool the cylinders, no need for the extra fuel)
Check out Bushcur's solution (where this is only applicable b/c it is understood that you have a buschur flash, which is conservitavely rich enough so that the coolant enrichment is not needed):Buschur "chip" results


Also, if anyone can find that value of resistance or voltage for.. say... 180 degrees F, and could tell me what simple electrical devices to use (reisistors, etc.) to keep the sensor value from exceeding it, I would greatly appreciate it. <I'm terrible @ circuits>

propellerhead

iTrader: (3)

According to the AEM EMS coolant sensor calibration table for the Evo:

V-------degF
0-------217
0.16---207
0.31---196
0.47---185
0.62---174
0.78---163
0.94---154
1.09---145
1.25---136
1.40---127
1.56---118
1.72---109
1.87---102
2.03---93
2.18---86
2.34---79

Richard L

iTrader: (5)

The values I have given you is not accurate but shows how the coolant sensors behaves in general.

I am not sure if the fuel will change that quickly to allow you to trim the a/f ratio in a dynamic conditions. It appears that it work with bushur's engine, so it must work in yours?

I am happy to help you out with circuitry but you need to tell me what you want to do with the information propellerhead has kindly provided.

I welcome anyone put an effort in making a WI system themselves, I hope more people will do it - it is a very good concept and it has been proven many times over. Aquamist is expensive because we have to make virtually every part, each part has to be designed before feeding the cad/cam files into the machines. There are no short cut to it.

Just going back to the coolant signal converter, do you want the signal to be clamped, reduced or completely re-mapped?

Richard

PS propellerhead, thank you for the ECS data.

propellerhead

iTrader: (3)

No problem.

I think the idea is to clamp the resistance value to the equivalent of 190degF or thereabouts. This will trick the ECU and prevent it from dumping fuel as part of it's zealous anti-detonation "strategy".

This little trick is only recommended for folks who are doing short runs at the drag strip and are running on 100 octane (R+M/2) fuel.