PDA

View Full Version : Controling individual cylinder WI with PWM


Alfa-Vector
13-08-2014, 12:43 PM
I am planning a WI build on a ITB 4 cylinder with each inlet manifold runner having it's own jet. I like the aquamist system as they allow for PWM control of the nozzles, however I'm a little confused on the delivery method and can't seem to find the information anywhere. :(

What I would ultimately like to achieve is to run fully sequential fuel and water, although from the looks of things, I will have to run semi-sequential water injection.....but even the best approach to doing this eludes me.

From what I gather, water goes from the tank to the pump, followed by the WFS and FAV. In my case, there would be a 4 way manifold after the FAV and from the manifold to each cylinder. From my understanding, the FAV receives a signal pulse which activates and squirts all four nozzles at the same time? Is this correct? I find it a little peculiar if each intake port is receiving water spray every 90 degrees of crankshaft rotation.

I plan on not using the controller and relying on a third party engine management system with it's own water delivery table.

So I guess my question is this: to run semi-sequential would I need two FAVs and two manifolds; and four FAVs and four manifolds to run full sequential? Please forgive me if I've mad a stupid slip up and missed something bleedingly obvious.

Cheers

Richard L
13-08-2014, 05:26 PM
It will be difficult to implement a mutlipoint-sequential injection set up base on our existing components.

1. The FAV has to have quite low flow - around 500cc/min per manifold runner. 1300cc/min is our standard issue.

2. Detection an individual blocked jet require multiple flow sensors. Th=It is difficult to make a turbine base flow sensor that reads below 100cc/min.

3. Driving the FAV can be shared/parallel electrically with the fuel injectors if the coil impedance is matched.

Anyone of our system can receive a PWM signal form your ECU, frequency between 50-400Hz.

Water injection is rarely needed before boost. Once on boost, air speed is quite fast, distribution is not too much of a problem if the can be placed at the exit of a FMIC. Droplet distribution will even themselves out along the journey upstream.

If you are running a large amount of water, then consider a 5-jet setup, four at the runners and one at the exit of the IC. Due to the pressure differentials between the two banks, checkvalve jet is preferred.

Alfa-Vector
15-08-2014, 05:47 AM
Thank you Richard for such an in depth answer.
So I take it, it is going to be over-complicated, cluttered and not to mention expensive to attempt such a setup.
I do assume that the best approach from here would be to run a single bigger nozzle, say, in the piping between the air filter and airbox?

Alfa-Vector
15-08-2014, 06:59 AM
Oh...I forgot to mention that this is a NA motor....and based of my injector calculations, if I was to go down the individual cylinder path, each nozzle would have to flow just over 26cc/min at 10% mixture strength or 40cc/min at 15% mixture strength

Richard L
15-08-2014, 08:13 AM
At those low dosage, we cannot manage those low flow rates. In sequential mode, it will have to be even lower quantity per injection event.