water/meth injection

[ericjon262]

I've been racking my brain trying to come up with a solution to mount an A2W intercooler in practical manner on my car, and every solution brings just as many problems as it solves, and it got me thinking, how about water injection?

pros:

Compact, and easy to install.
can be controlled by the MS3
it works.
Overall, it is a simpler install
lighter than a complete A2W setup.

cons:

distribution problems with single point injection
can run out (A2W doesn't)

I feel like I could implement a system fairly easily, at which point my A2W headaches go away until I can come up with a true solution unless I decide just to roll with this.

I guess what I'm really looking for is a reason not to throw the A2W stuff on the shelf and use water meth instead. the system greatly reduces the number of headaches the A2W setup has been giving me.

[Shaun41178(2)]

I wouldnt do it. I am assuming e85 isnt available?

Meth works. Its just not for me. You have to tune for it. Which MS might have a failsafe built in if it doesnt activate?

If you tune your air fuel curve for it and its not there you will run lean unless MS has a safeguard of some sort.

Put an intercooler where the stock cat was. You should have plenty of room down there

[ericjon262]

I wouldnt do it. I am assuming e85 isnt available?

Meth works. Its just not for me. You have to tune for it. Which MS might have a failsafe built in if it doesnt activate?

If you tune your air fuel curve for it and its not there you will run lean unless MS has a safeguard of some sort.

Put an intercooler where the stock cat was. You should have plenty of room down there

E85 is to hit/miss, and of inconsistent quality. I do have a flex sensor that I am going to use, and will tune the car for varying alcohol concentrations, but I won't rely on it.

I haven't investigated the MS3's control aspect very much yet, but I have enough unused I/O's to add a table switch to restrict boost based on lack of injection.

so, in a perfect world, I would have that room available. at some point in time I thought it was a good idea to cut out the stock front cradle crossmember and install a 2x3 box beam further forward in it's place, so I don't have the same space available. I have been looking at modifying the beam to provide clearance forward of the transmission for the intercooler, but the modifications would be pretty extensive, and I'm not sure I want to cut a crossmember at the front transmission mount to add an intercooler, the idea hasn't been removed from the list yet though...

[pmbrunelle]

My gut feeling is to not immediately trust an intake manifold designed to carry air to distribute water/meth equally.

You're thinking of a spray-bar in a spacer type deal sandwiched with the throttle body? I think I would use a leaf blower to blow into the intake, and then spray liquid at the same time. Then, I would see if what's coming out of each runner looks the same. See if a fog is coming out of each runner, or if it's liquid that condensed on the walls, and is now dribbling out.

With a carburetor, you need an accelerator pump shot to compensate for the fuel wetting the intake manifold's surfaces, or else you would have a lean spot as you tip into the throttle.

I think that water/meth from a single point would be subject to the same phenomenon; spray, wet the walls, and then once the walls are loaded with water/meth, then the cylinders will suck in water/meth.

A bunch of uncertainty... I decided to mostly sidestep these questions by going with port injection.

********************************************************************************

MS does indeed have an input that could be used to signal a water/meth failure. I think it can cut the ignition upon receiving the signal.

********************************************************************************

Unfortunately, I haven't actually used my water injection system yet, so I don't any lessons to share at the moment. I'm busy sorting out various issues on low boost (wastegate spring pressure), so no need for the water now.

Except for that time when I inadvertently turned on the spray at idle, and I flooded the engine some 5 km away from home... I did manage to restart though.

Another time (at home) I wasn't sure if I sprayed while the engine was off. So I didn't want to restart, fearing hydrolock. So I pulled the spark plugs in order to safely clear the cylinders of any water that may have accumulated.

********************************************************************************

If you haven't read NACA Report 756, read it! It's free.

The report warns of the possibility of contaminating your engine oil with water.

When you start using water/meth, you should probably crack open your oil drain plug regularly to see if water comes out.

********************************************************************************

Can't you restrict boost based on the ethanol content of your fuel?

[ericjon262]

My gut feeling is to not immediately trust an intake manifold designed to carry air to distribute water/meth equally.

Agreed, and this was actually one of my first arguments against it. That being said, the 3500 plenum offers lots of space to plumb lines over each port entry to provide even, reliable distribution.

You're thinking of a spray-bar in a spacer type deal sandwiched with the throttle body? I think I would use a leaf blower to blow into the intake, and then spray liquid at the same time. Then, I would see if what's coming out of each runner looks the same. See if a fog is coming out of each runner, or if it's liquid that condensed on the walls, and is now dribbling out.

With a carburetor, you need an accelerator pump shot to compensate for the fuel wetting the intake manifold's surfaces, or else you would have a lean spot as you tip into the throttle.

I think that water/meth from a single point would be subject to the same phenomenon; spray, wet the walls, and then once the walls are loaded with water/meth, then the cylinders will suck in water/meth.

A bunch of uncertainty... I decided to mostly sidestep these questions by going with port injection.

one thought was to inject as close as possible to the heat source, in this case, the compressor. I could very easily inject in the compressor volute, giving ample time to vaporize the mixture, or do a quasi port injection in the top of the plenum. your test idea is a valid idea, and worth trying, the only part it lacks is the temperature aspect which would also help to vaporize the mix. how significant is the temperature of the intake? I'm not sure, I have yet to instrument such a test.

MS does indeed have an input that could be used to signal a water/meth failure. I think it can cut the ignition upon receiving the signal.

I figured I could wire two switches in parallel to ground, one to provide a tank low level, and one to provide a system low pressure, either one trips and it uses the table switch input to adjust boost pressure and timing according to the lack of charge cooling.

Unfortunately, I haven't actually used my water injection system yet, so I don't any lessons to share at the moment. I'm busy sorting out various issues on low boost (wastegate spring pressure), so no need for the water now.

Except for that time when I inadvertently turned on the spray at idle, and I flooded the engine some 5 km away from home... I did manage to restart though.

Another time (at home) I wasn't sure if I sprayed while the engine was off. So I didn't want to restart, fearing hydrolock. So I pulled the spark plugs in order to safely clear the cylinders of any water that may have accumulated.

bummer, I was hoping you might have some input on how well the control aspect with the MS3 worked. that was smart not to try and start without knowing the engine was clear. I will add though, I've seen attempts to start engines with fuel filled cylinders(Roadkill's Rotsun comes to mind), they typically don't have enough umpf from the starter to do damage, but it's definitely a better safe than sorry situation

If you haven't read NACA Report 756, read it! It's free.

The report warns of the possibility of contaminating your engine oil with water.

When you start using water/meth, you should probably crack open your oil drain plug regularly to see if water comes out.

I'll have to check that out! fun fact, water in oil is part of why modern engines have higher operating temperatures, when the oil is over 212F, the water boils out of it!

Can't you restrict boost based on the ethanol content of your fuel?

I'm 99% sure that is a tuning feature available in the MS3 pro, but I haven't explored that too far yet. I finally got the M12 CAN bus connector I was waiting on, so hopefully I can get the MS3 powered up and connected to the laptop so I can update firmware and start working on some of the basic parameters.

[draven]

What AW intercooler locations are you having issues with?

[The Dark Side of Will]

one thought was to inject as close as possible to the heat source, in this case, the compressor. I could very easily inject in the compressor volute, giving ample time to vaporize the mixture, or do a quasi port injection in the top of the plenum. your test idea is a valid idea, and worth trying, the only part it lacks is the temperature aspect which would also help to vaporize the mix. how significant is the temperature of the intake? I'm not sure, I have yet to instrument such a test.

You actually do NOT want to vaporize the water/meth mixture. The same mass displaces way more air/oxygen when vaporized than when in droplets.

[Shaun41178(2)]

Methonol carries its own oxygen atom.

[ericjon262]

What AW intercooler locations are you having issues with?

Honestly, all of them. I have even considered reworking the hotside to reposition the turbo for more clearance. I've built a car that is a giant pile of headaches, every good idea had a unseen fault waiting in the wings, thankfully with the latest revisions to the build I've been able to remove a good many of them and make several lasting improvements.

You actually do NOT want to vaporize the water/meth mixture. The same mass displaces way more air/oxygen when vaporized than when in droplets.

I was under the impression that the whole point was for the mixture to cool the charge via evaporation? am I wrong? or is the idea that it evaporates in the chamber instead?

[pmbrunelle]

one thought was to inject as close as possible to the heat source, in this case, the compressor. I could very easily inject in the compressor volute, giving ample time to vaporize the mixture, or do a quasi port injection in the top of the plenum. your test idea is a valid idea, and worth trying, the only part it lacks is the temperature aspect which would also help to vaporize the mix. how significant is the temperature of the intake? I'm not sure, I have yet to instrument such a test.

What seems to be a semi-common concept is to use a spraybar pointing towards the compressor's nut in the middle. As long as the mist is fine enough, you won't erode the compressor blades. The compressor helps to make a uniform mixture.

Also, you can use boost pressure to pressurize your water/meth tank; no pump needed. You would probably need an air-atomising nozzle to do the job.

An ECU-controlled solenoid valve could be used to implement RPM scaling.

I figured I could wire two switches in parallel to ground, one to provide a tank low level, and one to provide a system low pressure, either one trips and it uses the table switch input to adjust boost pressure and timing according to the lack of charge cooling.

Never use a closed switch contact to indicate an unsafe condition. If a wire breaks, you'll never know it.

Instead, use an open switch contact to indicate an unsafe condition. For multiple contacts, wire them in series. A wire breakage here will cause the system to revert to limp-mode.

I'm 99% sure that is a tuning feature available in the MS3 pro, but I haven't explored that too far yet.

You don't need an ECU to install TunerStudio and start exploring the various features. It is possible to work on a tune "Offline", and then upload it to the car. I have attached a screenshot of the water injection settings in MS3 1.4.0. I did it from my couch. My car is not in the living room.

You actually do NOT want to vaporize the water/meth mixture. The same mass displaces way more air/oxygen when vaporized than when in droplets.

I think there's some optimum size: small enough to be carried by the airflow, but with a big enough mass to surface area ratio to not evaporate instantly.

Methonol carries its own oxygen atom.

Yup! And as long as the alcohol remains liquid, it won't take up much volume, leaving more available volume to be stuffed with air.

[pmbrunelle]

I forgot the screenshot!

[ericjon262]

for some crazy reason I got it in my head that I couldn't do anything in TS without first connecting the MS3 to it... I have no idea why. That's a really important catch on the safety circuit, thanks for that. I think I would either route would result in me using a pump, the higher pressure would allow for a finer mist, and coupled with a pressure regulator, a higher level of control and tunability.

[draven]

The two best aw install locations on a turbo fiero were matt hawkins, who mounted his under the driver decklid vent and fieromadman from this board, who mounted his in place of the catalytic converter.

Do you have any pics of your engine mounted up in the bay?

[ericjon262]

The two best aw install locations on a turbo fiero were matt hawkins, who mounted his under the driver decklid vent and fieromadman from this board, who mounted his in place of the catalytic converter.

Do you have any pics of your engine mounted up in the bay?

The turbo interferes with the decklid grill install, and the crossmember interferes with the cat converter location. to mount the IC I'll probably have to cut/clearance.

[Honest Don]

Is it a different than stock crossmember? How big is your A2W?

If the turbo hits the vent, maybe remove it and hang the turbo out in the breeze? Sometimes you have to just lean into these things.

[ericjon262]

Is it a different than stock crossmember? How big is your A2W?

If the turbo hits the vent, maybe remove it and hang the turbo out in the breeze? Sometimes you have to just lean into these things.

Turbo is nowhere near hitting the vent, but you can see the air intake to the turbo will be occupying the space under the vent. The crossmember is moved forward, closer to the firewall by 4-6", which make for a very narrow gap for the IC to squeeze.

How big is the A2W core? about 6.5-7" thick, 13" long and about 10" tall. but mounting it is also only half of the problem, I also have to get charge piping and water lines too and from.

[Honest Don]

I could see how moving the crossmember complicates things. What does your intercooler look like? Would tilting it help?



It’s not the most elegant solution, but some of this:

https://www.jegs.com/i/JEGS/555/15101/10002/-1

Along with compression elbows and other fittings from the hardware store might help with working out the water lines

[The Dark Side of Will]

Methonol carries its own oxygen atom.

Not as much as air.

I was under the impression that the whole point was for the mixture to cool the charge via evaporation? am I wrong? or is the idea that it evaporates in the chamber instead?

There's an optimum... some amount of cooling in the manifold makes the charge denser, which must be balanced with displacing air with vapor to get the greatest mass of air into the chamber. Then the rest of the water/meth can evaporate on the compression stroke and during combustion to prevent detonation.

[draven]

can you reverse your intake manifold and place the TB on the passenger side then use a log/barrel type AW intercooler where the cat would be and route charge pipe up to the passenger side.

See fieromadman's setup..

[pmbrunelle]

can you reverse your intake manifold and place the TB on the passenger side\

You could run a dizzy if you did this!

How about putting your intercooler in the cavity ahead of the left rear wheelwell. Would you be able to fit it there? I assume that you would need flex hoses / rubber to accommodate the engine movement.