Water Injection Setup with MS3+X - What should i buy?

[MWPau]

Hi all,

Quick engine intro:
Toyota 5S-GTE (2.2L 4cyl, all forged) with ~10.5:1 CR, mild cams and smallish factory ct20b turbo.
Will be run on E85 normally, but needs to run on 98 or ever 95 octane unleaded if needed.
To handle the high CR+boost when on unleaded, i know that water (plus maybe meth) injection will be very beneficial, if not crucial.

My question is... what type of water injection setup/components should i run?
Of course, i would like it to be controlled by the MS3+MS3X.

I do want a variable rate of injection so i can move between 10:1 (fuel:water) to 4:1 depending on engine load (probably tied to injector pulsewidth, or close to it).
Though if that proves difficult to achieve, then 2/3 different sized nozzles with solenoids controlling on/off flow to each.

So what similar setups have you guys run, and/or what can you recommend?

Thanks in advance!

[Tjabo]

Over the last couple of years I have been messing around with a variety of water injection components. I've been controlling them with a CoolingMist CMGS controller, but always with the mindset of eventually controlling them with the MS3X.

After much research and pondering, I have come to the conclusion that the best bet is to simply call Jeff Howerton at Howerton Engineering:

http://howertonengineering.com/

If it were me making the purchases at this point, I'd go with his standard Aquatec pump, and his standard fast acting valve, which I think is this one:

http://www.howertonengineering.net/prod ... Valve.html

I spoke with him a while ago about putting together a basic kit for MS3 users, and he was very positive about it -- a GREAT guy, too!

One note, I've found that unless you plumb in a multi-port system, it's not necessarily to go with a large volume of water/methanol solution, in that it is difficult to get good distribution through an intake manifold intended for dry flow (wow, big epiphany there! ). Also, the water/methanol creates a tremendous amount of exhaust volume increase, probably largely from the steam. This volume can be used to assist in spooling a largish turbo if you initiate injection from a low MAP value around 100kPa or even less with MS3 (With the CMGS controller, I have only been able to trigger mine as low as +1psi).

The downside of the additional exhaust volume is that it exacerbates boost creep, so it is necessary to have a properly angled exit pipe to the wastegate.

Hope that helps!

Thad

[prof315]

A friend a I are looking into building an MS compatible progressive controller. Essentially it's going to br a high current FET with a driver/controller that will allow you to PWM control the pump voltage and thus pressure (which is how most standalone progressive controllers work). We should begin testing within a week or so.

[Tjabo]

Why would you do that when you can run it natively from the MS3X?

Also, the ability to PWM the nozzle with a consistent pressure as you can from the MS3X is vastly more mappable with good atomization. Confused I am! LoL...

[prof315]

Why would you do that when you can run it natively from the MS3X?

Also, the ability to PWM the nozzle with a consistent pressure as you can from the MS3X is vastly more mappable with good atomization. Confused I am! LoL...

It's going to be native from the MS3/3X . Use the pump output to ground a relay coil and provide power to the pump and PWM the driver/hi current FET with PT4 ( as if it was the solenoid) the driver/FET package PWMs pump ground to control pressure. We're building the driver/FET because there really aren't any circuits on an MS3/3X able to cope with the current demands of the pump (and the noise it is sure to generate).

I agree that PWMing the nozzle at a constant pressure is a better way of doing things. What we're designing is going to cost far less than a water/methanol compatible fast solenoid. Also water/meth compatible solenoids are scarce. The one you linked is the only one I have been able to find.

This way of doing things will be cheaper and rather generic. It should also work for doing PWM control of electric water pumps and fuel pumps.

[Tjabo]

That last point is definitely a bonus! How cheaply do you think you'll be able to build the motor PWM controller then?

My only real reservation with this is that when running a good bit of methanol you really need some good mapability to cope with big temperature swings, and I don't know if volume adjustment via motor speed control can really do it without sacrificing a lot of atomization at the low end. Temp swings have been quite a pain for me, but right now I have no ability to compensate for it at all... I'll be very interested to see how this works out!

[prof315]

That last point is definitely a bonus! How cheaply do you think you'll be able to build the motor PWM controller then?

My only real reservation with this is that when running a good bit of methanol you really need some good mapability to cope with big temperature swings, and I don't know if volume adjustment via motor speed control can really do it without sacrificing a lot of atomization at the low end. Temp swings have been quite a pain for me, but right now I have no ability to compensate for it at all... I'll be very interested to see how this works out!

Probably under $50 including the aluminum case. As far as atomization goes, I'm going to use AIS nozzles which are _supposed_ to atomize fully at 50psi and up. My Cooling Mist Smart Injection system has a 250psi pump so that should give me a respectable range. Needless to say i plan on doing some faily serious testing off car including volume tests.

[Tjabo]

FWIW, I've done quite a bit of pressure testing with my Coolingmist "250psi pump" setup, and the thing couldn't break 200psi with anything other than the smallest nozzle. I had to bump up to some higher RPM (and higher amperage) pumps, as well as some high lift pump cams to really get the pressure range I wanted. I think it seemed as if I was going to need to plan for about 15 amps if I was going to try to have enough pump to use a pump speed PWM system.


But then I found out about Howerton's Fast Acting Valve.....

[prof315]

The FET we're doing initial testing with can handle 75A! Yeah I know overkill but it's free and the same mfr makes a 25A FET for cheap.

[Tjabo]

I was told I should shoot for twice the capacity of what I expected the motor to pull, so that sounds GREAT!

[MWPau]

So, something like:

Tank -> Filter -> Pump -> Pressure Accumulator -> Pressure Sw (Turn on Pump) -> Highspeed Valve -> Nozzle

?

[prof315]

Quick update: It's taken a little longer than we hoped to get a prototype built but it's done and real testing starts today.

Some changes, the FET driver works on an input > 4.5V. It works really well with 12V. So the initial testing will use the MS3X PK0 (tacho) pin with the pull up installed. This same driver can control 2 FETs so we may set things up with an optional second output. We're also going to get some further input on the design from a true expert, (my friend Steve who is doing the actual electronics design works for NASA) the power systems electronics team leader who is working on the next gen Space Shuttle! I'm hoping we can invert the control input to ground based so it will be more friendly with most MS outputs.

BTW maybe this thread should be moved to the expansion boards forum.

Jeff

[prof315]

Testing of the high current driver proto was a total failure. The circuit needs to be active low (which I suspected all along) because the active high outputs available (tacho w/JP3 installed and PT4) don't float low enough to release the driver.

Just for giggles I wired the pump ground to pin 36( D14 BIP373) and tried using it. It works, but more than 2 minutes at 90+% DC overloads the BIP and it shuts down. It _should_ be OK for running on the street but we are still going to get a working high current (25+ amps) driver together.

[franksidebike]

Hallo
can the V3.0 Bank1/2 circuit a AME Water Pump handel?
in the AME Video she says the Pumpe have 6amp!
frank

[prof315]

Hallo
can the V3.0 Bank1/2 circuit a AME Water Pump handel?
in the AME Video she says the Pumpe have 6amp!
frank

I'm actually using the mainboard INJ1 driver to run my pump successfully. See my thread here: viewtopic.php?f=125&t=42036 AEM is full of it though....that pump is an Aquatec like everybody else's and will draw 10-12A

[franksidebike]

thanks jeff thats the info i was locking for
the msextra forum is too big for my brain
frank

[franksidebike]

Hi
My mainboard INJ1 driver is broken
Yesterday i have a run for 30s (?) with the AEM wather pump was on! Now she is olways on
Before i have a bench test with stim made for 60s and the INJ1 circuit is littel warm!
What can i make?
frank

[franksidebike]

Hi
I have Q3, D5 and D6 changed and all is good
frank