ky7

Just bouncing around a few ideas. One disadvantage of our PowerFC is there are no additional 0-5v or PWM outputs for driving additional injectors etc.

I'm looking for a setup where I don't have to size nozzels and at the same time enable water injection to follow a specific fixed water / fuel ratio based on PWM signal from the primary injectors. I also want to be able to adjust the the actual flow rate / ratio on the fly without having to size different nozzels...

I like the solution Vari-cool Controller from Coolingmist, specifically because of the following statement on their site...

Now, I want to provide this Vari-Cool controller with a 0-5 signal representing primary injector duty cycle. Aquamist have their 2d system, which requires you to size your nozzel for proper flow / ratio. Included with the 2d is the FIA2 unit, which amplifies the PWM signal and converts it to 0-5v. I'm thinking this unit or similar could be used to send to the Vari-Cool controller. The water injection system would be activated at a certain boost level, maintaining a specific fixed water / fuel ratio based on injector duty cycle.

What do you guys think...??

sdminus

I have this controller.

The problem with it is the 0-5 volt source. If you use the map sensor you can only meter the water in via the amount of boost you run.

On a car that builds boost fast its no good. IMO get the aquamist kit. I wish i had

Scott

The coolingmist statement is incorrect (factually) There is noway you can get the correct amount of water in to achieve good atomisation with a large jet on a low pump duty cycle. The pump would need to flow between 60psi and 250psi to even get close. You also need to keep the pressure upto 60psi to achieve atomisation on the jets.

Happy to answer questions

Howard Coleman

iTrader: (6)

i was initially interested in aquamist's FIA2 unit as our motors use less fuel after peak torque while running a constant boost. i figured the FIA2 would key off injector duty and reduce the AI to the motor after peak torque.

well, not exactly.

you have to pick as to whether you take the signal from the primary or secondary injector.

we have no injector duty cycle signal that will work.

wouldn't it be great if it could be derived w the new FC Tune?

if you have watched your primary and secondary duty cycles you will realize they interelate back and forth and neither would not work as a source of total fuel into the motor.

until you find a way to actually derive a duty cycle current the FIA2 will not work.

i don't know what the problem is, if any, w the Coolingmist controller but i run a progressive controller off the 3 bar GM MAP sensor and expect no issues as to tune as
there are over 1300 Alkycontrol systems in use, most w progressive controllers.

i do agree that the ultimate controller would key off injector duty cycle but you can have most of what you need now.

howard coleman

ky7

sdminus, thanks for the advise. I'm thinking the 2d system from Aquamist with the FIA2 unit is the route to take, but...

I've seen guys splicing into the front pri injector wire coming from the ecu to hook the PWM signal to the FIA2. You are saying this is not accurate / will not work correctly as it does not represent overall injector duty?

Please elaborate..

I'm not sure what you are getting at.. the injector duty is easily logged. If you are looking for some sort of 0-5v signal from the datalogit / fc-tune box - I would not trust this as there would be delays due to i2c comms relay from the ecu to the box...

RX794

You can use the injector duty cycle signal from the ECU, it's shown on the instructions of this AVC-R on page 18 of the manual itself or page 19 on Adobe when you're looking at it on the net: http://www.apexi-usa.com/pdfWiring/8.pdf It's diagram Z3-b, you'll see it noted as injector.

ky7

That's terminal 4W - Injector (front primary) as per FSM for US models.

So that brings me back to the issue howard has raised...

Pardon the pun, but are we getting our wires crossed?


As a side note.. since my car is 96-2 jap import, I'm referring to ECU Wiring Diagram Z4-a. The connectors don't look familiar to me. As far as I can remember my connectors are the same as Z3-b (will check when I get home).

sdminus

The problem is. My stock twin build boost by about 1.2 psi every 100ms so it takes around a second and a bit to go from vacum to about 12psi. In this second the controller has to switch on the unprimed pump get it upto 60psi of pressure and adjust the mixture at 100ms intervals. There is no way it can mange this. The reaction time is just to tight ?

also the jets do not perform below 60 psi. and the pump wont flow about 150psi. This gives you a narrow band to play with.

Scott

The injector duty idea would help at high end to stop it from over saturating the mixture

Howard Coleman

iTrader: (6)

“I've seen guys splicing into the front pri injector wire coming from the ecu to hook the PWM signal to the FIA2. You are saying this is not accurate / will not work correctly as it does not represent overall injector duty?”

it would be accurate relating to the primaries.

No, it would not work correctly. If you choose to use the primary injector signal what happens when the secondaries kick in and the primary signal ramps down? This is very clearly shown on the Power FC Map watches and realtime charts.



“I'm not sure what you are getting at.. the injector duty is easily logged.”

The duty cycle is reported as a number not as voltage or current. If you look at the notepad on the power fc all you will see is the %. There is no 0-5 voltage signal.


“If you are looking for some sort of 0-5v signal from the datalogit / fc-tune box - I would not trust this as there would be delays due to i2c comms relay from the ecu to the box...”

I am no electronics expert but I do a lot of tuning. I, for instance, log two digital egts and they are spot on as to readings. I see no lag. If there was a duty cycle signal it would seem that it would generate at the same time as the injector ontime which seems to my untutored eyes to be instantaneous.


“You can use the injector duty cycle signal from the ECU, it's shown on the instructions of this AVC-R on page 18 of the manual itself or page 19 on Adobe when you're looking at it on the net: http://www.apexi-usa.com/pdfWiring/8.pdf It's diagram Z3-b, you'll see it noted as injector.”

Yes I know there is an injector port. I run an AVCR and am familiar the fact it reports IDC. My guess is it functions just like the Power FC and just reports a number…..


I wish to stress I do not consider myself expert in this area. I would love it if someone were to be able to extract a usable injector duty cycle signal. My original point for the post was to discount the idea of hooking into the primaries or secondaries alone to run AI. Since they cycle in a manner relating to each other you won’t be able to derive anything approximating the overall IDC.

howard coleman

ky7

I see what you are saying..

The fact is however, that the duty cycle is not a signal, rather it is data in a packet, which is communicated either to your commander unit or to your box over i2c and then to your laptop over serial. This will inevitably result in timing delays, although tiny, it is still not "realtime". For instance, the anomaly a lot of people have observed regarding logging with the datalogit - when logging basic and advanced, there is a discrepancy between reported values, e.g. basic rpm value and advanced rpm value. That is because these individual packets are streaming in sequence and there is a time delay, however in the case of datalogit, it simply uses the msec time the first packet requests were sent per line in the log.


So what if it "reports" just a number. I don't know much about the AVCR, but if it only requires hooking into the pri injector wire from the ecu and can report duty cycle correctly, then it is extapolating the duty value based soley on pri inj signal. To me this would suggest that using the pri inj signal should be correct with the FIA2. Your thoughts..?

twokrx7

iTrader: (1)

Once you hit transition where both injectors are working they ramp up together. If you use the pressure switch to arm the 2d system, by the time the boost triggers the system you should be past transition and either injector signal should work to provide a signal to the aquamist system.

Another possible solution for those that run constant boost setting is to ratio water to rpm. Trigger the system at 14 psi and then find a signal or generate one with some simple electronics to feed rpm to the 2d, then map the 2d to provide the amount of water you want by rpm since we know about how much fuel is being delivered under full boost conditions at all rpm. Not perfect but should get you close, mostly what is lost is the ability to play with boost settings on the fly once you are tuned.

ky7

Thanks twokrx7.. I was thinking along these lines and I think this answers my questions. Particularly because my full boost would be .9 bar and I want to engage WI after a certain boost level - lets say .5 bar. It's past transition point for the injectors at this stage, so tapping the FIA2 to the primary injector signal is probably the better option. Then its just down to sizing the nozzle for the correct ratio I guess...

sdminus

once again kyle a top answer.......

Scott