Too Much AI?
 

Recently added the AEM AI kit with two 250cc nozzles located on the charge pipe between the outlet of the IC and the TB elbow. Running a BW SXE 300 with 62mm compressor at ~20 PSI on CA 91 pump gas and distilled water through the WI system. I have the Sakebomb AEM ignition upgrade kit.

Lost ~30 WHP on the dyno with WI vs. no WI. 455 WHP vs ~425 WHP. Tuner noted significant ignition breakup just as the turbo gets into the powerband. It's as if the water injection is putting out the ignition spark. Also noticed that there was so much water that the pressure sensor for the electronic boost gauge also became clogged with water intermittently during dyno pulls. The AEM controller was set to start water injection at ~10 PSI and reach maximum flow at ~30 PSI.

Is it possible that 500cc of distilled water is too much for my setup? We're going to take out one of the injectors and try again at the dyno with a single 250cc injector. Also converting from wasted spark to direct fire with new spark plugs.

Thoughts?

What type of ignition system are you running?

The amount of water doesn't sound horrible, but it's dependent on the amount of fuel you're injecting. I think the sweet spot is supposed to be 12.5 - 15% of the volume (not mass) of your fueling injected for any given time.

Run a few calculations as far as your fuel consumption during a WOT run (it's easy, don't give me bullshit about not knowing how), then compare it to your AI Water injection. I know Shainic did a nice graph of the AEM Pump Output, so I'd suggest finding that on here and adjusting duty cycles as necessary.

I have the Sakebomb ignition kit. Was running with wasted spark and am converting to direct spark.

The IGN-1A Coils need more time to cycle than what rotaries allow them (in wasted spark). Waste Spark was probably getting them toasty with little output. Direct Fire, they'll really be in their element. Great kit, great company to deal with.

2700 CC of fuel under peak load

So that's over 18% Water Injection, which is definitely not going to work well with the IGN-1A coils in a wasted spark configuration. Once those suckers are firing Direct Fire, they'll blow through that with ease.

Even still, I think you want to go down to ONE injector, probably 300-400cc in size (PEAK!!), or limit the PWM of the pump to something less (so you don't have to buy new parts) and keep both injectors as is. Really simple, right?

Orrrr, inject more fuel. :D

Going to try a single 250cc injector (AEM nozzles come in either 250cc or 500cc sizes). Returning to dyno tomorrow with the new settings.

Going with a single 250cc injector did yield better performance. Set the AEM controller to start spraying at 4 PSI with full flow set for 25 PSI. See this thread for more info:

https://www.rx7club.com/single-turbo...1108778/page2/

Glad to hear it! :) I think you'll be doing best with pure water with 250cc, or 50/50 water meth with 500cc.

For consideration...

According to the AEM manual, for 450 HP @ 2700 fuel flow and your boost, you'll want 826 AI flow. I am using 500 for a much milder build (300 HP)


https://cimg7.ibsrv.net/gimg/www.rx7...a461953f5a.jpg

It could be you are opening the AI too early. The manual says the start psi should be 1/4 full boost. If your full boost is 25 psi, that's 6 psi. Not sure 4 to 6 would make that big a difference tho


https://cimg5.ibsrv.net/gimg/www.rx7...881a1e1d81.jpg

Another consideration is the integrated check valve. Maybe that's faulty. You clould be dripping water in your intake when the system's off. Pull the injector and see if drips at all.

Mixing with methanol is a plus because methanol burns and cools the intake charge.

He's doing purely water, which is 10-15% of total mass of fuel. Water/Meth in 50/50 configuration is MUCH higher 20-25% is an average there.

AEM Clearly States W/M = Water/Meth Injection.

I wouldn't necessarily characterize W/M as MUCH higher than pure water. If AEM is calling for 826 for W/M, think 500 would be good (w/o doing a bunch of math) and 250 for pure water might be too small (obviously I'm no expert tho).

How are the nozzles orientated? I personally think they are coming on too early or they are leaking (or combination), but again i'm no expert.

You may also want to try one nozzle. I'm not sure if there's any pattern benefit for having two (unless you have two controllers that could operate the two nozzles independently, a smaller one coming on at lower pressure then a larger one coming on at higher pressure)

I dont understand how "nozzle requirement" changes based on boost; when the fuel flow is static (I assume they are talking about max fuel flow).