Understanding Engine Knock/Detonation

[turbophil]

I've been meaning to post something about this for a while. I elected to enjoy some time in front of the computer during our part-time holiday hours to put together a little piece on this interesting topic...

In the following link (should be able to click the picture, if not, just go to boefab.com for the link), I used our dyno to illustrate detonation and knock. Within the link, I also added one of the best summaries on the topic I've ever read.

Perhaps a good thread to continue to discuss the subject as well?

Anyway, hope you find the link helpful...

Phil

[addertooth]

Awesome article, very well worth the read. Almost makes me want to go read my plugs today. You had to stay on top of this stuff with some of the air-cooled hemi-head performance engine builds I used to do.

[turbophil]

Glad you liked it. I should probably add more pics.... Allen's summary is long, but worth while.

I hope the dyno traces help to show normal harmonics compared to destructive ones. Perhaps of interest, the detonation indicated in the traces were not severe enough at that load level to be heard by the naked ear. I have ear phones attached to a knock sensor that allow me to hear detonation that would otherwise not be able to be heard. That said, even with the ear phones, the digital readout and dyno traces "rules all" since my ears have proven time and time again to not be the world's most precise instruments

Cheers,

Phil

[addertooth]

What would also be useful would be ignition timing curve overlaid with the AFR, then the total function of AFR/Timing could be compared with detonation.

[turbophil]

Originally Posted by addertooth

What would also be useful would be ignition timing curve overlaid with the AFR, then the total function of AFR/Timing could be compared with detonation.

I think I'll avoid using specifics when it comes to timing. Invariably, someone will read a bit about X timing detonating, and Y timing not, apply Y timing to their setup, blow up their motor with a finger pointing at this thread. Since there are plenty of variables across plenty of different setups, will just keep it to generalities.

The primary variable in these pulls was timing. You can see AFR varied a bit, but in the critical areas the AFR was all within an acceptable range for this motor and fuel (pump gas).

The timing on dark green and aqua blue pulls was the most advanced (too much for pump gas on this setup) and was introduced around 6,000 RPM. Up to about 6,000 RPM, all the the tunes had similar timing curves with only the IAT corrections (post blower) changing the applied timing a bit. You can see that around 6500 the aggressive timing in those pulls started to detonate. Had I increased the load on the rollers, they probably would have detonated audibly. Even hitting a 60 on the scale as the aqua blue trace did is not something you would pick up with your ears at that RPM and load.

All the other pulls had several degrees less timing with the last two having the least timing and that showed up in the power traces with lower output.

As for AFR- shoot for 11s in the critical areas of the curve (upper registers and torque) on these relatively mild blower/turbo applications on pump gas. You're not leaving much if any power on the table and still getting some cooling from the fuel. Yes, you'll perhaps burn more gas than you may *need* to, but that's a good/safe generic number for many. Short of having the ability, patience, and tools to really dial in your setup for lower fuel consumption, I'd recommending sticking with being a bit overly rich rather than finding a perfect spot in the 12.X's at 8,000+ RPMs and Y psi boost with stock compression. Keep in mind that all the fuel in the world won't keep detonation at bay.

Cheers!

Phil

[Westrock]

Be nice to know what it sounds like, on the 2ZZ. When mine started running like crap it would make a sound at full throttle that sounded like BB's in a coffee can being rolled along.

[ERP]

It also sounds like an electrical storm on the 2ZZ, a crackling buzzing sound. If you want to hear it run your A/F in the 14s at WOT.

[turbophil]

Yes, you'll generally know knock when it's happening and you can hear it. The tougher part is knowing knock when you can't hear it!

[ChronosWS]

Great article. If you can find examples, could you throw pictures in the article showing the various indications of failure (land failures, four-corners scuffing, spark plug destruction, etc.?) They would add a lot, I think.