Water Injection

[prj]

Yes, but get the AN steel line upgrade as well.
Don't use the push-to-connect fittings.

[Rusty]

For sure

[20Vcqdriver]

Was lucky getting my Aquamist system off e-bay, Originally bought by a young lad for a Shitron Saxo and clearly had no idea how to make best use of it so I got my PWM kit for £127. Bargain...
Took 6 months of patient waiting and watching to get it, but worth it.

[prj]

Was lucky getting my Aquamist system off e-bay, Originally bought by a young lad for a Shitron Saxo and clearly had no idea how to make best use of it so I got my PWM kit for £127. Bargain...
Took 6 months of patient waiting and watching to get it, but worth it.

I hate you.

[Billman]

I see that Aquamist also changed to diafragm type pumps and not piston any more. That PWM kit is nice but I think with a solenoid it is easy to map it nicely with controllers for singe extra injector or VEMS (for those that use it).

[prj]

You don't really need a solenoid.
You can just PWM the pump and have a check valve in place. You need a 20 amp capable FET.

[Billman]

These diafragm type pumps need some ms to get into pressure. If I understand it correct with the solenoid the pressure will be always high.

Maybe an electronic (mappable) boost controller could work also by mapping DC vs boost vs RPM??

[Thuppu]

With Aquamist the high fequency solenoid is adjusting the mix.

[prj]

These diafragm type pumps need some ms to get into pressure. If I understand it correct with the solenoid the pressure will be always high.

No, you understand wrong.

First of all there is a check valve before meth delivery. So if the pressure goes below 1 bar it will not inject. This will stop any kind of dripping problems.

Second, the PWM is pretty fast. 30hz. It is basically similar like varying voltage to the pump, except it's better.
There is no "takes a couple ms to get into pressure" problem here. It's something in your head, same way like boost control pulses the N75 valve, and the boost is nice and steady. It just goes too fast.

It does not matter really if you pulse the pump or pulse the valve. Your nozzle is not a fuel injector. So you are just using the PWM on the line, but the output function of that PWM is still pressure. You just control it differently.

In both cases the pressure differential at the nozzle is a function of of PWM.
The only difference is if you control it at the pump, or you control it with a solenoid, but the end result is exactly the same...

Also, it can't be any different. A fuel injector's dead time is 0.8 msec for EV14's.
Now imagine, if you can only inject every 0.8msec, then turn it off for 0.8msec again.
The problem will be that you will have a very uneven delivery of meth, that's why you have continuous injection at a certain pressure, whether you PWM the pump or PWM the solenoid.
It's the same way for example K-Jetronic works, where all injectors are open and there is a pressure regulator there which gets adjusted continuously. You are doing the same thing, since you have manifold injection.

If you were to fit 5 meth injectors into your manifold, then that would be a completely different story. But with 1 or 2 nozzles, while in theory PWM-ing the valve may seem like a better solution - in reality there is not any difference. The only real difference is, that you need a 20 amp capable FET to PWM the pump, while you can do less than 1 amp to the solenoid. That's it.

[quattro84]

Agree with PRJ, increase pressure differencial over meth injector/nozzle = higher flow and no need to ponder about on/off nor dead time behaviour. Not worth the hassle unless you want precise monitoring of injected mass. (Would be fun to log @ WOT how long it takes after the first injection event before it's starts cooling the charged air let alone becomes stable)
Agree with the push connectors, they suck and will get leaky over time especially after refitting, BTDT on newer rallycars which use IC sprinklers.

Does anyone have a clue on the meth/water consumption compared to fuel as a ratio?

[prj]

Does anyone have a clue on the meth/water consumption compared to fuel as a ratio?

You can't take it directly like that, because it really depends on how much WOT you use.
Not much point to spray meth/water below a certain threshold from a performance point of view, so you could go through 5L on two laps at the track, while the same could last you two tanks of fuel in day to day traffic.

It also depends on nozzle size, etc.

[quattro84]

Ofcourse, there is no fixed ratio and i wasn't expected to get such an answer but i have no clue hence the question. I am after a max rate to determine injector size.

Reason for asking is because i have access to an EV14-bodied injector with unique 3-tipped nose used for dosing well, some h2o based solution @ 10bar constant for fine evaporation. Max flow rate? Must find spec sheets but i think i heared 4kg/h so i suspect that'll be more than sufficient.

[prj]

Pump is 150-200 PSI.
For some cooling a single 1mm diameter nozzle is usually used.

If you want to run a lot, and tune for it + remove fueling by the ECU not to go too rich, two 1mm nozzles are used.

I am sure you can figure it out from this data
It's true for both the 2.2T and the 2.7BiT engines.