Individual runner WI + intercooler outlet?
 

I'm thinking about setting up WI for individual runners and the intercooler outlet. Thinking about it, on the plus side there are the benefits of air chilling and also "direct port" water to combat detonation. On the negative side, I think this could be difficult to tune and I'm also not sure how to set it up properly.

I'm thinking a Y splitter with one fork going to the IC and the other going to the IM would work. Fork the IM Y again and then fork the fork for 4 nozzles for the runners. Use same sized .5 nozzles for the runners. Maybe use a .8 nozzle for the IC?

For tuning, I think it might be best to tune both the IC and IM WI separately to get food flow targets, then plug em both up and mess with the maps or the nozzles to get things right.

What do you guys think?

Useless...
Dont complicate things more than you have to.

Why useless?

i know of a couple guys who did individual jets on their MKV GTI.. they saw no diffrence between that and using just a single port. the effort to go about setting up the extra ports was a waste of time.. they said it wouldnt have been a total loss if there was any gain.

i know its not on a miata but they were doing the same thing.

Interesting. I want to do individual runner though to make sure each cylinder gets the same amount of water.

ya thats what im talking about.. they stuck a nozzle on the under side of the intake manifold in each one of the runner.. ill see if i can dig up some pics

cold side of the intercooler should get the cylinders damn near identical, the air is turbulent enough coming out of the IC, and there's plenty of distance before the manifold. A lot of guys aim jets at the throttle plate, and dont have any problems whatsoever

im still looking for another set up which showed how he used only T's for distrubiting the meth. but anyway heres one

Attachment 194481
Attachment 194482
http://i558.photobucket.com/albums/s...a/972f2376.jpg

Woahhhh WTF is THAT. I was going to use Ts or Ys but that shit looks much sturdier.

You can get distribution manifolds like that from a variety of suppliers. DevilsOwn has a few of their own, which accommodate standard 1/8 NPT fittings: http://www.alcohol-injection.com/dir...ifold-129.html

If you have a drill press and some scrap aluminum bar stock lying around, they're not difficult to make. While you're at it, you can switch from press-in fittings to screw-type compression fittings if you want. Check your local ACE hardware for the fittings- they're the same ones you use to connect icemakers and in-counter filtered water dispensers.

Your going to need individual ones. Because one nozzle WILL NOT cool all the cylinders easily. LULZ

Thank you! What is your opinion about my original question Joe?

??????????????? WTF does this mean? I'm guessing you're being sarcastic and don't know much on the topic? :hustler:

BTW FWIW here's some post I found on an Evo forum straight from Mr Aquamist himself recommending my idea.

http://forums.evolutionm.net/4187845-post21.html

If there were a god though, he's not it so I am still interested in hearing other people's opinions.

And here is a post from mr snow performance recommending this also

http://forums.evolutionm.net/7776022-post14.html

My opinion is that I had six Mojitos with dinner, and it's a miracle I was able to walk the four blocks back to the hotel, much less log into the forum and post a link to the DO manifold. You don't want to be listening to anything I might have to say about nozzle sizing right now.

I don't really understand the idea behind having it in BOTH places.

Just makes no sense to me.

Either do 1 big nozzle or 1 small per cylinder. Why create a clusterfuck of potential failure points?

Yeah, that. Pretend I wrote what 18psi wrote, only with big words like "elucidate".

For what its worth, I strongly question the notion that evaporative cooling prior to the combustion chamber is the principle method of action of WI insofar as its anti-detonant qualities are concerned. I haven't done any original research on the topic, but a lot of what I've read of the old NACA studies suggests that the presence of water mist in the combustion chamber endzone during the compression cycle is where the real jello's at.

IOW, further away from the chamber inlet != better.

1 big nozzle. no need to complicate it... most meth/water setups are all kind of a wild ass guess anyways exactly how much you are spraying.

Sarcasm. Branching from arguing with Tom.

The idea is that the IC nuzzle is far enough away from the cylinder head that when I am slamming 3000000pzi into my engine it will e-cool the charge bitchfucker on it's long trip to the cylinder.

The IR zonnles are too close to the head for e-cooling and the water will get slammed into the cylinder before it evaporates in the charge bitchfucker. The unevaporated water will then release it's shitload of latent heat energy when it evaporates when the spark goes off. That shit slows the flamefront of the combustion event, cools off the piston, cylinders, valves, all that shit.

My idea (in my mind) would supply

a) nice ambient temperature air to engine
b) in-cylinder e-cooling

As for failiure points that is taken care of by failsafe- check valve, flow senzor, tank level sensor. If any nuzzle fails, flow will take a shit.

the real advantage of water is getting it into the cylinders to lower EGT....it's high specific heat and all that.

Nice ambient air temp to engine.. are you planning on having the IAT after that nozzle or something for lower indicated temps?

You're injecting the same shit into the bitchfucker from different distances. I highly doubt there's a point to that.

But I have to say its just my opinion and I'm no expert so I'd love for someone that knows the ins and outs of this shit to prove me wrong.

How much power are you trying to get out of this engine?

Yes, IC bung is on the IC outlet.

IAT sensor is about 20 inches away. Long pipes :|

Direct water injection. Be a G.

http://www.knvts.nl/S&W%20archief/Di...%20vessels.pdf

It actually looks like more for diesels and low emissions. Your idea for the direct portish wi made me think of this.

from what I've been reading about water/meth and iat placement.. you want the IAT before any nozzles- reason being if water hits the IAT... its basically going to read the water temp not your air temp. also the water evaporating on the sensor is going to make it read even cooler, the issue here is i am assuming your tuning uses IAT to determine how much fuel to inject. Yeah you can just offset it a little but ideally you of course want open loop fueling to be as accurate as possible.

I was just looking into this debate a good bit today... good discussion on the Megasquirt forum and some good reading I found on evo forums as well.

Aaron, most dealers will say to put the nozzles before the AIT sensor, and use the sensor to tune for the water meth. At least that is what I get from the SRt forums.

To the OP, I think that the direct port will be the best way to go, but having a nozzle before that will make sure that even if one of the nozzles gets clogged, that that one cylinder is still getting flow. Depending on if you go with a PPS or aquamist style system, you will need at least one more check valve or solenoid because the direct port nozzles will have vacuum that the IC nozzle won't, and the IC nozzle will have gravity feeding that the DP nozzles won't.

If you are going to make your own distribution block, give it 5 ports and feed from the middle. T before the block and feed your IC nozzle. Check valve on the port, and check valve on the IC nozzle.

Absolutely no gain in performance between direct port and 1 nozzle.

WI's real advantage is knock suppression, if you want a cooler charge buy a A/A IC.

Another thing worth considering.
4 x nozzles will mean you will have to select very small nozzles, somewhere around 0.4mm, in my experience these nozzles tend to block up a lot more regular than say a 0.7mm.

Now think of this.
If you have 1 out of the 4 nozzles that has blocked or partially blocked, it's going to be extremely difficult to detect.

If on the other hand you have just 1 big nozzle and that blocks up any flow sensor worth it's salt will instantly pick up on this.

My advice is to mount just 1 nozzle on the intake.



Cheers
Mark

That sure is strange....

I always assumed that knock suppression was the real advantage of cooling the intake charge, and that both the A/A IC and Water Injection (With its remarkably high specific and latent heat constants) both perform basically the same function.

I suppose I am a complete fucking moron. :nxsmile:

Interesting reading on where to mount the nozzle in relation to the AIT sensor, there doesnt seem to be any consensus on before/after the temp sensor. It also seems like a lot of these guys are using a speed density system as opposed to a MAP sensor for measuring air charge. Is a speed density system the same as our MAF? Seems like they're worried about screwing up their speed density calculation (which, if I understand correct, is MUCH more important) instead of their air intake temp calculation.

Speed/Density IIRC is a MAP sensor type management. The ECU's tables are based on RPM (speed) and manifold pressure/vacuum (density). Both systems use an AIT sensor.

Speed density is basically MAP.. what the megasquirt defaults to etc. MAF is totally different and in my other car... what matters is the flow velocity and temp at the MAF sensor- we have another temp sensor in the manifold (boosted air temperature), but this is only really used for the ECU to determine which load maps to use as far as I can tell. In speed density the air temp plays a pretty huge role from what I can tell in fueling. Open loop idle conditions I will get my AFRs perfect, then the next morning they will be super rich... or vise versa, IAT gets really hot and everything leans out.

I also found a good thread on the MS forums but I can't find it again.. basically the conclusion was put the IAT/AIT before any nozzles etc, that way you are getting the AIR temp (which is used to determine how much oxygen content it has per whatever)..not the fluid temp which will result in the ve tables being as accurate as possible.