NefMoto

I asked this on audizine, but I figured it deserves a spot over here as well.

This is a spin off of the knock voltage thread... http://nefariousmotorsports.com/forum/index.php?topic=444.0

What do you all think are safe numbers for correction factors? (on oem knock control maps)

I have heard under 10, under 7 and some think that no correction factor is acceptable.  What do you all think?


Would really love for expert tuners to share some insite on this. On other forums like Evo, Dsm and Subaru forums, the tuners don't seem to mind sharing this to help the advancement of the community.  I really hope we can get as much of this info as public as possible. Many more of us are tuning our own cars, or logging and talking with our tuners. Any info would help!

Also some thoughts on modifying the knock detection methods, is this safe when done within limits? Do any tuners do this? On nefmoto there has been talk of reducing the knock sensitivity.



Here are a couple of mine for reference (stock knock control methods):
3rd gear
(http://i97.photobucket.com/albums/l214/boomerw/3rd_timing_1.png)

2nd gear
(http://i97.photobucket.com/albums/l214/boomerw/2nd_timing_1.png)

multi gear pull
(http://i97.photobucket.com/albums/l214/boomerw/MultiGear_1_2_3-IgnitionRetardCyl1IgnitionRetardCyl2IgnitionRetardCyl3IgnitionRetardCyl4IgnitionRetardCyl5IgnitionRetardCyl6andIntakeAirTemp.png)

(this thread is not just about my personal logs, but it helps show what I am running, id love to see other peoples as well)

Thanks in advance!

I've heard from multiple people here on the forum about advancing until you get ~6* timing pull.

Uhm looking at one of my 3rd gear pulls it seems that my ignition retard is negative and measured in KW according to ECUxPlot... am I logging the wrong values here?

(http://nerd.hu/ign_small.png)

No, it should be negative, as that is how much is being subtracted.

The thing you are aiming for is the most timing in the majority of conditions.  Looking at CF in isolation won't tell you a great deal. 

Rick

The confusion here lies in the fact that ECUx flips the sign. ME7Logger does not.


Agreed. But as a rule of thumb, if you are logging in best case conditions (good gas, low IATs, etc.) you should see almost no corrections (small CFs).

If you are logging with bad gas, high IATs etc. you may be leaving timing on the table if your CFs are too close to zero (small CFs)

dwkrz is naturally negative IIRC and added to zwgru for zwbas. I think I have it flipped in my *.ecu file because I prefer to graph it as a positive number.

I find that pulling timing will usually result in more advanced overall timing. For example with random made up numbers (don't do the math and tell me they aren't factors of 0.75), say zwbas is 15* BTDC and wkrma is -6, if I pull 1.5* wkrma will drop to -2 and zwbas will rise to 17.5* BTDC.

Edit: just wanted to clarify that I used wkrma for simplicity and that I know that it's unsigned...before anyone trys to pick

My experience as well. ME7 is very conservative; it generally pulls more timing (with stock KR maps, of course) than it has to... so you have a lot of timing correction under perfect conditions, you are most definitely leaving timing on the table.

Stock thresholds are massively conservative.  Det cans and good ears are your friend here.  Once i've finished calibrating these, I get no timing pull in "good" conditions.

Had a car on a runway recently (1.7 miles - don't know if that is considered long in the US? Brake distance incl.) and i was pulling no timing and zero intervention from EGT sensor.  Car (RS4 - not the best aero) hit 187 in this distance which is pretty amazing on stock turbos.  We did multiple runs and she just wanted more.  Beat lots of other marques on the day :)

Rick

So potentially pulling back timing to reduce the correction factors will run more total timing because the ecu pulls too much?

I kind of hinted at this with modifying the way the knock voltage is filtered, is it safe tone down the filters and limit the amont of timing the ECU pulls due to each "knock event"?

Is there a general consensus on the "sweet spot" of CF's? 

If you're not leaving timing on the table, and the ECU isn't over-correcting by pulling too much timing, your CFs should be _______ .

My guess would be -3 to -4.5?

The meth kit had everything to do with that.
I run more than 1.4 bar to redline and can run 24 degrees of timing while at it with meth.
Turn that off and it's a very different story.