I suspect your engine may be close to what they call the 'incipient' detonation phase. That means its close to almost detonating, but you can't hear it. Best power happens when the engine is just on the brink of knocking. 8% or so is a likely gain if you can rasie the normal advance enough at part throttle, and not have it knock at wide open throttle. Adding water hurts the peak power level, but can improve the part throttle crispness. Pontiac Paul, my mate, says he only uses water injection so he can run 42 degrees total advance on his 455 Formula drag car without wrecking the cast pistons. Off the mark launches, with a 2.73:1 diff in a 4200 pound Pontiac, were right on 2 seconds for the 60 foot time becasue he was able to avoid detonation. He can never leave at 4800 rpm, always about 3000 to avoid wheelspin. The best torque happend with water injection on, but the overall mph was lower. With H20, he'd get 13.8 at 98, or 13.7 at 99 without. So if ignition is adjusted to suit, H20 improves torque, but hurts peak power.
In a high altitude area or with a planed head or any engine that has polished chambers and ports is more likely to detonate than a stock engine. Perhaps the cam is quite aggressive, because in a normal 300, detonation only happens when the engine runs a high compression ratio.
If the block is decked or the normal 80 thou shortfall of the piston from the block is removed, the detonation should be reduced. (Normally, with the dish in the piston, the register is 220 thou down including the piston trough)If your running a lot of intial advance, then water injection may help the engine in moderate engine speed, high load situations.
I keep going back to the things that cause detonation, and it has to be a combination of those 22 things.
1) flame travel due to plug placement (its most often reduced with hi-po pistons),
2) valve shrouding (often increased, but is mitigated by gas flowing with a famous SuperFlow flow bench!),
3) piston dome masking (often, the dome is relief cut to suit flame travel and valve opening).
4) Effective compression (cam related octane demand). Doesn't always go up with increased compression, as the whole reason for it is usually to raise opening duration to increase effective compression. When a racer bolts on a 320 degree cam, he can loose a huge amount of effective compression, which reduces the octane demand.
5) lastly, quench is often increased as at the 10 to 14:1 level, people go to alloy heads with closed chambers or welded iron or alloy production heads.
6) intake heating
7) fuel distribution
8) head material (iron, alloy)
9 Inertial Ramming (Carburation, efi, vee engines have better thermodynamic properties)
10) piston deck impinging , flat or recessed.
11) blue printed to 0.1 c/r or C/R balanced on all cylinders accept the detonation prone ones. (If they are reduced by half a point, you can go up another 0.5 points on the other cylinders)
12)the charted mixture motion of incoming air fuel mix
13) Cold Cranking compression pressure influences the compression an engine can stand.
There are other factors such as road load due to
14)tires,
15)wind drag,
16) frontal area,
17) air density,
18) wind,
19) the slope of the road,
20) weight influence,
21) torque converter,
22) overall gearing steps.
The en gineers set peak advance from tests with all these loads combined.
Whew!
My book of the Nebraska Tractor Tests uses gallons of fuel per horsepower hour so just wanted to be certain.
!
