One nozzle for each aux. port runner ....

Interesting idea. How are you going to confirm whether you'll be able to safely run more boost? Monitor exhaust temps, check for pulled timing, use knock headphones, use virtual dyno to check whether power is increasing proportionally to boost?

For those interested, see post #12 posted by TeamRX8 on rx-8 injectors:
How does the fuel injection system work?
It looks like injector activation is based on load, and has no direct correlation to aux port opening. If Brettus' mod is effective it would indicate that fuel injector placement in the aux port, and injector activation triggered by port opening may have been a sensible design decision.

I'm not looking at it so much as a way of increasing boost ...... hoping it will make more power on the same boost. If that is the case then yes , it may well have helped make power on an N/A engine if Mazda had put injectors in those runners. Should be able to check it using virtual dyno.
Actually ............ In theory ..... if all settings are kept the same I should be able to tell just by seeing how many g/s it flows . Before vs after.

Did a test tonight .Set it up with a boost switch plus an rpm activated relay . Initially I thought there was no difference but I realised all the w/m was getting sucked out of the line post check valve so........... that is causing a delay in activation . Maybe something good happening once it does activate but need to resolve the issue before confirming.
Really need another check valve so i can mount one on top of each nozzle ...that might solve it .....maybe.

While I am no expert, I believe even if the water hasn't vaporized by time the motor begins compressing the charge that it will be quickly absorbed once those rotors begin squeezing that mixture down. Since the MSP is 10:1 compression and at 15 psi added (turbo) you are more like 20:1 dynamic compression (294 psi). Compressing the mixture 20 fold will generate so much heat that the water will vaporize well before it comes time to fire the plugs.. Of course, you could always add too much water. I use the same principle to inject directly into the compressor blades of the turbo and the engine itself.

Yeah , good point . Possibly 100% methanol may work better ... but will stick to 50/50 for the sake of safety.

I found 70 /30 to work well when I was running it

where is your tank? Unless you get a leak and spray it on the manifold you should be fine.

Using the windscreen washer tank ... it's really only just testing the theory at this stage so I guess trying a higher concentration should be safe for that.

I wouldn't do that... especially in a Turbo 8.

I was looking to find a picture to show you why.... ...the alcohol tends to evaporate at a much lower temp than the water.... and when it gets a nice concentration over the hot manifold bad things happen.... ask me how I know....
https://cimg7.ibsrv.net/gimg/www.rx8...597a57db80.jpg


Where did you mount your pump? I know it's a long way to the trunk..... but it is a lot cooler back there.
I put in a 3 gal fuel cell for the Meth feed tank and mounted the pump under the trunk floor on the right wheel well area.

Might give up on the idea anyway .

Virtual dyno shows no gains . All I can see from the logs is a gain after the boost cut activates at 7250.
It seems like the cooling effect takes a while to do anything ...heat on the tube maybe?

mass airflow vs rpm with w/m in red without in blue . Have done several runs and see same result each time.

https://cimg8.ibsrv.net/gimg/www.rx8...b229e7b05b.png

I don't think you should really see any gains from adding methanol/water to a system (most lose power). The power comes from additional timing and boost.

On my 11:1 compression mustang I would get tons of knock running more than 6-7 lbs of boost on 93 octane. I changed my head-gasket about 5 times in a span of 18 months. I added a water/methanol injection system with two 5 GPH (315 ml/min) nozzles right before the throttle body, and run about 9 psi on a Precision 7675 and haven't had knock or changed a head-gasket since. Its more of a safety cushion.

The Renesis is somewhat unusual in that the APV runners have no fuel injection in them .... Was thinking this could be a way to improve flow though that runner only and gain a little. Any ideas on why I'm seeing a slight gain after boost cut ?

You can see afr take a dive (as w/m activates) about half way along the log and boost cut just after vert. blue line.
https://cimg6.ibsrv.net/gimg/www.rx8...ed43dea6f8.png
Vs log ran just after with no w/m
https://cimg4.ibsrv.net/gimg/www.rx8...5430c8eceb.png

Actually ...... just noticed that just before and after boost cut .... the boost is almost 1 psi more which does account for the extra g/s ...but why ?????

Is your rev limiter cut fuel, ignition, or both?

Rev limiter closes the throttle .... but that's not what we are looking at there . Boost cut just turns off boost controller and opens the wastegate to spring pressure

Oh gotcha, you turn the duty cycle to 0 above 7,200 or so. So on spring pressure alone you are getting more flow because you are getting slightly more boost.

So you are trying to figure out why your are making slightly more boost on waste-gate pressure alone (while injecting). The richer/denser mixture while injecting meth/water could be putting more volume through the turbine speeding up the turbo enough for a bit more boost. It may just be insignificant. I can do back to back runs in my car with the same tune and get different logs. :dunno:

Whenever I've made changes to my setup that result in increased boost at the same WG spring setting, I've observed increased MAF readings and improved exhaust manifold to intake manifold pressure ratio (exhaust pressure may have increased slightly, but the ratio is better). I'd guess that if you measured your exhaust to boost ratio, it would probably confirm improved efficiency. Must be caused by lower exhaust gas temperatures, reducing exhaust back pressure at any given MAF reading under boost, meaning the engine can breath in more air.

I'd run back to back logs with and without w/m injection using WG spring only, so there's no boost cut causing confusion in the logs. At the same time I'd log exhaust manifold back pressure to get a clear Exhaust to boost ratio for comparison.

I'd probably gather another set of logs with duty cycle set to run highest boost without hitting the boost cut. This may show more noticeable differences with and without injection, as the airflow is deeper into the compressor efficiency island.

Another test that takes more effort would be to lock the aux port open the whole time, and rig up the injection so you can switch it on manually. Then gather logs with and without injection switched on. This way you'd get an entire pull with injection without worrying about any lag or w/m pooling as the aux port opens. This would give you more direction on whether the trigger system for the injection is effective.

I was thinking along the same lines ...... must be something like that.
The setup pretty consistent from one run to the next ..... I'm def. seeing that change.

Since you already have nozzles, and are on your way to Nobels price in research, you might want to try Hydrogen Peroxide(H2O2) in approx 30% concentration. If memory serves me right, it transit into H2O+O at 4-500°C, this requires energy, lowering the temperature, and on top of that you have added an extra O in liquid form(= very dense = just about no of the normal air is displaced = more air at ignition). I'm not sure if the alloys will be able to take H2O2, as it's very reactive with some.

Did some more w/m injection experimentation with various nozzle sizes and locations .... sadly I didn't find any extra power.
However, I have been giving it some stick on some windy roads and absolutely loving the feel of this setup running at the high 300s level . Power onset is smooth and very controllable.I don't need to rev it very high to have fun. Handling and brakes manage the power with ease. I've never driven a supercar but to me this car feels like one when in that environment.

:icon_tup: Nice!