For all of you out there who tune your own ECUs, be it Ecutek, Link, Motec whatever, how much difference does it REALLY make (in terms of BHP, or torque, I suppose) to tune the ignition advance right to the limit of detonation compared to being a couple of degrees more conservative in the interests of engine life?

Thinking logically, it must make a reasonable difference because the Subaru factory strategy for knock control is quite sophisticated, keeping the timing 'on the limit' according to octane, air temps etc. Subaru have obviously gone to some effort, so it can't be for nothing. But I am hard pressed to feel a 'seat of the pants' difference between close to the edge tuning and quite conservative tuning, especially when on boost. Low vacuum and low boost are another matter, but 3 or 4 degrees less than where it was tuned on the dyno at any boost more than about 10psi doesn't seem to make any difference. It was tuned on the dyno with det cans until it detted, then brought back a degree or two until repeatable clean pulls through the rev range were possible.

I'm basically nervous about things like hotter air temps and or coolant temps, variances in fuel octane etc bringing on detonation, because the air temp compensation for timing seems to be an educated guess by the tuner at best. But all this talk about J&S Safeguards and other ways to determine det have me worried that I may be throwing performance away by taking it too far from the 'edge of no return'. Is the Subaru factory strategy just there to prevent knock rather than optimise performance?

Any comments/experience/figures greatly appreciated.

PS I have headers which, I have been told, tend to allow far more spark advance than the standard manifold. In fact my tuner said I will start to lose power by advancing the timing too far before it will ping, so maybe you don't need it that close to the point of det if you have headers?



[Edited by BenD - 11/26/2003 12:58:51 AM]

Bump.

I know there are some hardcore tuners out there who could answer this.... JB? BR?

I can't help on the advance/power ration thing (although I do remember that it has been talked about before, but a long time ago & I can't think of any search that wouldn't bring up masses of crud at the same time :rolleyes::), but yes you certainly want it as close to det point as safely possible, irrespective of headers! I'd have thought that it would be possible that aftermarket headers could actually allow less advance because of the increased VE they would provide, although of course this doesn't mean they produce less power because of less advance, as the advance is merely particular to that engine setup :)
Not sure if that made sense? :D

Yes, tried a search and came up with all sorts of threads about things like 'timing belts' etc...

I was under the impression that the stock headers were so restrictive that they didn't allow efficient scavenging of all the exhaust gas, therefore increasing the heat of the incoming charge which is one of the causes of detonation. I could be wrong here, but this is my basic understanding of why increasing the volumetric efficiency of the engine could allow more spark advance.

[Edited by BenD - 11/26/2003 9:47:35 AM]

Mainly, yup. First bit is bang-on. But it can be a tad more complicated... heat might be left in the chamber because the exhaust hasn't been entirely removed, but I think that exhaust gas slows the flame front down. Which sort of compensates.
You might have better luck on the 22B forum (http://bbs.22b.com/) as it gets a tad more technical in there :).
If you can cope with the huge, long, documentary fashion of his explanations, look out for the drivel that 'pat' spouts - I imagine he's gone into this in ridiculous detail somewhere :rolleyes::D

am not great at such things, but on the dyno with the addition of nf we were able to add 2/3 more degrees of spark advance.

Have a look here, iirc the NF was added between the third and fourth graphs so you can see where the difference.

Torque went from 526lbft to 560 something and power went from 548 to 568bhp.

If that is the difference made by just adding NF (maximum concentration) to super unleaded you can imagine what difference optimax or race gas would have made.

I'll tell you, it would have cracked the block sooner! :rolleyes:

OK, that's decided it for me. I don't trust my abilities to listen for det, even with headphones and an amplifier with a piezo microphone. All I can hear is loads of tappet noise. I will probably fork out for a Knocklink now, although I am dubious about the accuracy of these as well. Any other solutions? Hmmm maybe I need some sort of active knock control, which means a new ecu.....$$$$$......(grumble, mutter mutter). Probably not worth it for a few more KW.

[Edited by BenD - 11/27/2003 12:01:24 AM]

Ben, cans & experience are the best way. I wouldn't trust a Knocklink, useful though they are, to give me a definate diagnostic

completely agree with Jlang,

Cans are easy to make and having heard sparkle through them on the dyno, I am sure you would be able to hear det a mile off.