Hi,

Have planned for water/meth injection for a long time and recently I finally clicked "Order". Got the AEM 30-3015 pump today and started the try-out


Plan is to build my own controller since I am already doing quite lot of work with micro controllers.

Did this simple test just to see how to control the pump flow and its seems to be very easy. Voltage close to 0 gives a slow dripping from the output hose while 10V gives many litres per minute. Next I will sort out the exact relationship between Voltage input and the output flow of water.

But how much water/meth should I inject into my.4.0 Xkr? It has got the 10% smaller upper pulley, K&N air filter and full cat back exhaust from Milltek

I understand that the volume to inject should increase with ambient temp, high rpm and so on.

 

Probably worth reviewing the various sources such as AEM, Snow, Aquamist, etc.
Quick answer = since I have aquamist, I followed their guidelines, which you can see if you download a user manual from their website - for 50/50 mixture they suggest 1.5 cc/min per HP, or 15%-20% of fuel, also you can review flow versus pressure charts for their nozzles.
Back of the envelope math (hope its right - few glasses of wine)
Since 4.0 SC has 8*360cc injectors (at ~45 psi?), at 80% IDC (if that's what is used at max HP - no idea if it is...) and 20% WM to fuel ratio this would be 360*0.8*8*0.2 = 461 cc/min max flow.. Or stock 370 HP, then 370*1.5 = 555 cc/min max flow.
So somewhere around that kind of range....
With a progressive system you can meter the flow referencing suitable parameters like injector duty cycle and/or positive boost, etc., so it somewhat tracks the engine power.
Not sure how well you could do progressive by varying the pressure from the pump, since low pressures might cause poor atomization of the mixture, depending on nozzle maybe you could get 20-40% range?? Obviously easier but more expensive if you use a pulse width modulation controlled valve, since pressure is kept high.
However, the simple on/off systems (e.g. boost threshold triggered) seem to work well based on their popularity - I assume you choose a compromise/average value near the peak engine power, but have not used one of those systems.
Also you have to decide where to inject, pre or post SC or both, and that's another variable - different 'ambient' pressures, % split, warranty on SC, etc.

 

Thanks Chirpy, good for me to have some figures to lean on when trying to sort things out in my head.

Interesting way to do it with a pressure tank(?) after the pump, and then control the flow into the intake from the tank instead of directly from the pump. Should be faster response, yes.

The documentation of the AEM kits, as far as I can see, are feeding the liquid directly from pump into intake as shiwn below.



Thats what I plan to do as well. Im thinking of setting a series of thresholds from ambient air, so that really cold ambient air will prevent the system from triggering at all. Medium levels of ambient air temp will allow for water/meth spray at high levels of throttle position/rpm. Higher levels of ambient air temp will allow for injection earlier.

To be done with an arduino reading info from the can bus, and a voltage regulator to send the correct voltage to the pump. As you wrote, probably no point of running the pump very slow, it wont give the pressure to yield a spray

 

Did some more testing with all three "nozzles" that were included in the V3 nozzle kit. Or to be more correct, there is one nozzle but three different "pintles" that you put inside the nozzle and that decides how much fluid that will be sprayed. 250 cc, 500 cc and 1000 cc pintles are included in the kit.

I have now tested the flow with all three pintles and with a range of input voltages 6-.12 volts.

First of all; to run the pump at voltages below 6 V is of no good. The mist will not form properly, so always run this pump at 6 V minimum.

Flows are not linear with increased voltage input as you can tell from the table below. Increased voltages closer to 12 V will not yield much difference in sprayed volume. With the 1000 cc pintle I found no difference in sprayed volume between 11 V and 12 V.

The test was not done with scientific precision, so take the figures with a pinch of salt. In my case, I will go for the 500 cc pintle and probably run the pump at voltages between 6 and 11 depending on conditions.

Sprayed volume per minute:

 

Hi,

Varying the voltage to control flow for this application is fraught with insurmountable issues.

One should establish a tight voltage range of operation and then use a pulse width modulation interface to control the flow rate.

It is not that difficult.

I still have several ‘kits’ from years back and will dig one out over the holidays and will forward it to you if you are interested.

wj

 

Hello Higgins,

My M/W System is from Snow...I've been using the Snow System with the Round Controller for years.....I use my Windshield Washer Bottle as my M/W source. Bought a little coupling from them and drilled a hole in the bottom of the Bottle.....Ran the hose to the Waterproof Pump and from there to the Inlet Neck above the T/B, I used the next to smallest injector they had. Don't remember the Number as Snow is now NOX out of Wichita Falls, TX. Look them up and they have all this information on the Snow part of their Website. I hooked up the pressure to the hose that comes out of the S/C....Had to "T" it off and ran the Hose to the Little Controller that I put on my Windshield post....I can set the Controller to come on at a certain pressure and then even more juice at a Higher Pressure. It really works great! I have a Guy here in Houston that set the System up for me.. I will try to send you some photos. As a Plus....I can still use the Windshield Washer....The mixture of M/W is 50/50 and works just fine as Washer Fluid. The Controller is set to come on @ 9 lbs. And then again at 19 lbs. Good Luck! Contact me anytime!

Billy Clyde @ Cinco Ranch

Snow Performance Controller mounted on Windshield Post

 

I understand that some pumps, maybe older pumps in General, were designed to run at a given voltage. In this case I can see that the mist forms perfectly from 6v and upwards. No bigger drips at all