For the second run we turned the water injection on and left the timing at 32 degrees total. We took care to ensure that the intake temps stabilized between every run to get consistent data. With the water injection on and the same ignition timing, the NSX made 332hp and 252 lb-ft of torque to the wheels. Which equates to a 2hp and 1 torque loss over the non-injected run. Our intake temps were at 170 degrees when the water injection kicked in, and by redline, the intake temps had dropped to 159 degrees (compared to 192 with the injection off).
With a 33-degree drop in intake temps at under boost, the detonation threshold should be moved up a bit, allowing us to advance the ignition timing to take advantage of the cooler air. So, Kuroyama bumped the ignition timing up to 37 degrees total and let her rip. This time around, the NSX put down 349hp and 258 lb-ft of torque to the wheels, with intake temps again hovering around 160 degrees under boost.
In the end, the water injection netted us 15hp and 6 lb-ft of torque. While that is a significant gain, the gains could have undoubtedly been much more substantial on a higher boost car. Because Law's car is only running around 5psi of boost, the supercharger isn't even generating enough heat to benefit greatly from the system. Even without an intercooler, the car was running just as much timing as a naturally-aspirated car. With the timing set at 37 degrees with the water injection on and no detonation (stock naturally-aspirated NSXs run around 32 degrees), the gains would likely be much greater is we started with a car that had to retard the timing and add fuel to keep away detonation. This was simply not the case with Law's car.
Nonetheless, the water injection was doing its job and lowering intake temps substantially. Because the air temp sensor is at the bottom of the intake manifold in this application, the intake air temp sensor reacts more slowly and can register a drop in temps only after the water has cooled the heatsoaked supercharger. In fact, after just one pull on the dyno with the water off, the supercharger was too hot to touch, but immediately after a run with the water injection, it was only lukewarm. This decrease in intake temps not only added power, but will likely reduce a great deal of stress on the engine.
In the end, we didn't really make the power gains we would have liked to see (or the gains Snow Performance says they typically see), but it wasn't really any fault of the water injection. The application we chose to install the kit on just wasn't generating enough heat to really benefit. Now, if Law ever gets the urge to up the boost, he could benefit much, much more. But nonetheless, even on an unlikely candidate like this car, water injection can still be a huge benefit, giving it extra power while giving it that extra degree of safety at the same time.
Now all we have to do is find a more "typical" application that runs retarded timing and rich air/fuel mixtures to keep detonation at bay (which isn't hard with the 91 octane swill we get 'round these parts). Then we'll see what it can do for the right car.