[Td-d]

Originally Posted by Fast_Freddy

AFAIK the ECU pretty much advances timing until knock is detected at which point timing is retarded until knocking stops at which point timing is advanced again until knock is detected and so on. How else could they achieve ideal spark timing across the wide range of fuel grades, altitude and weather variations the car will be subjected to?

Thanks for the link. Great info that I had not seen before. At a glance it looks as though IAM is much like long term fuel trim (LTFT) and FLKC is much like STFT.

No prob, there's a lot of information hidden on the RR forums over the years. Not really - it's a more reactive process - basically you have the base timing, and then the knock correction advance, compensated for with the various parameters (i.e. FBKC, FLKC, and IAM as a multiplier) and that's the timing. There are 3 knock control 'levers' - IAM is the so-called 'rough correction' mode, and will drop all timing in the KCA by a %. FLKC is learnt over time and stored in ROM defined 'cells' whilst FBKC is immediate.

Originally Posted by Fast_Freddy

You're not suggesting that our knock sensors or ECUs are cheaper versions than those found in Outbacks and Solaras? I don't agree with that. I do agree with your other points but would like to add that this is true for any tune, not just JV. Speaking of which, aren't there literally hundreds of people running JV tunes out there? Any reports of tune-related engine damage?

Not suggesting that, nope - I haven't seen the part numbers, but I would not be surprised if it is the same knock sensor that is used across most of the Subaru platforms. Just saying that consumer grade knock sensors have their limitations. On JV's tunes - really, I can't comment, and frankly am not interested in commenting - I avoid internet drama like the plague Besides, I'm half way across the world (literally 8000 miles!). I'm here for the techie info...

Originally Posted by Fast_Freddy

I may be wrong but what I learned in the 2AZFE tuning world was that the biggest problem with high KC was that the ECU pulled too much timing before adding it back which lost some power. It was best to minimize the amount of KC but not to eliminate it. The only case I know of where a N/A 2AZFE was destroyed by knock was when the knock sensor came loose on a bone stock engine and it self destructed due to the OEM timing strategy. My own turbo experiment yielded gains of 125whp/125ft lbs at 6psi with nothing but 20% larger fuel injectors and 93 fuel for tuning. No piggyhack, no ECU flash, nothing but a AFR sensor signal modifier for AFR enrichment while in closed loop. I raced that engine many times and daily drove it for over 40k miles without any issues.

I would agree that on an N/A platform it's less likely to as big a problem (I don't own a BRZ incidently, have an STI). On FI, due to the nature of Subaru's pistons (hypereutectic, not cast) and the compromises made for emissions over the years, on-going knock is bad news, and will likely take the rings out, if not worse. But I cannot (and won't) pretend to know much about the FA20 - there are smarter people out there! I would expect these engine to be more not less knock resistant given the direct injection into the chambers.