NOx climbs with increasing combustion temperatures. Therefore, in a nutshell, NOx does climb as spark timing is advanced (higher cylinder pressure = higher temp). NOx emissions and EGR are very interesting subjects. I have a summary/explanation posted on HybridZ, it can be found here. NO! Absolutely not! I don't mean to yell, just want to emphasize the point. I'll try to summarize why this is so. Once the mixture is ignited, it burns and releases its energy over a certain period of time. The whole point of having the spark timing advance is for the mixture burn time to match the engine speed. Burn time is essentially constant (depends on AFR, combustion chamber geometry, fuel composition, etc.), but engine speed is variable. Therefore, the burn must start earlier at higher engine speeds. There is an optimal point at each engine speed, where the burn must release a large part of its energy. In relation to crank angle, this is at about 15 degrees ATDC, if I remember right. If you ignite too late, the burn happens late, and your maximum energy is less (piston is dropping = bigger volume = less pressure) since it occurs at a less favorable crank angle, thus you make less power. If you ignite too early, your peak energy release occurs too early, when there is less leverage on the crankshaft (rod almost, if not completely vertical). Your goal when tuning a spark map is to find the "happy point", the technical term being Max Brake Torque (MBT) timing. The knock limit comes into play if your fuel does not match your engine (too low octane, too high compression, high intake temps, or combination thereof). The thing is, the knock limit can stop you before you reach MBT. However, if you reach MBT but have not reached the knock limit, then you can easily over-advance the timing. Therefore, just because you advance to just before the knock limit does not mean that you will be at MBT (the optimal spark advance for that specific engine speed/load). If you are interested in more, here is a great paper I just found on the topic: http://www.vehicular.isy.liu.se/Publications/MSc/06_EX_3809_JM.pdf Another great reference for those truly interested, is "Internal Combustion Engine Fundamentals" by Heywood. This is the ICE bible. You usually need to calibrate those sensors so that you're making sure they don't give false alarms. However, once you reach high enough rpm, these sensors tend to pick up a lot of noise and give false alarms, at least on our solid-rocker adjuster L-series engines.