|
|
|
|
Thread Tools | Display Modes |
05-03-2012, 06:33 PM | #1 |
Scooby Specialist
Member#: 159743
Join Date: Sep 2007
Location: Plymouth Ma
Vehicle:07 STI Limited TurboMike Tuned 2.35 |
Changing Rough Knock Correction to allow more Fine Learning
Im hitting a wall here where ive been tuning a 2011 NA and i want to try something new but i wanted opinions on if its a good idea or not.
now i have a decent timing map and the car is happy but there is this side of me that thinks the car could tune itself better. I say this meaning, Fine Correction, in a sub 1.0 or 16 IAM has the ability to add timing in spots with no knock as well as take it away when it is 1 or 16 it only can take away. So im thinking if i set the initial IAM at .875 and then turned the step value that moves the iam up or down to 0, would this be a good idea. Im not sure if im chasing 1 additional MPG or if this could lead to something bigger. I think of it as i could go into an NA and put on a much much safer map, Scale the Fine learning Load Columns since mine isnt seeing over 1.45 load ever and allow the Fine correction to add 10 degrees max.... Or if i should just be setting it and letting the rough correction do its thing... which only applys on this car over a .80 load since my timing advance table starts there... I hope i get some real feedback because this sounds like something i may try tonight on my ride home with the right feedback.
* Registered users of the site do not see these ads.
|
05-03-2012, 09:02 PM | #2 |
Scooby Specialist
Member#: 290109
Join Date: Aug 2011
Chapter/Region:
SCIC
Location: Fresno, CA
Vehicle:2005 WRX PSM |
off topic, and i apologize.. but i've been searching reasons for my IAM to keep dropping from 16 to 12. i have yet to be able to catch when it happens (with a log).. but we're kinda stumped. the only changes were an ebay TMIC and injectors. out of curiosity and without having to go too far off topic, any idea of why? just thought i'd ask since this thread came up as part of my search. thanks for reading.
|
05-03-2012, 10:48 PM | #3 |
Scooby Specialist
Member#: 274499
Join Date: Feb 2011
Chapter/Region:
NESIC
Location: RI
Vehicle:1994 5.8T Cobra 755rwhp @ 16psi |
Get rid of the eBay tmic.
|
05-03-2012, 10:55 PM | #4 |
Scooby Specialist
Member#: 290109
Join Date: Aug 2011
Chapter/Region:
SCIC
Location: Fresno, CA
Vehicle:2005 WRX PSM |
|
05-03-2012, 11:01 PM | #5 |
Scooby Specialist
Member#: 274499
Join Date: Feb 2011
Chapter/Region:
NESIC
Location: RI
Vehicle:1994 5.8T Cobra 755rwhp @ 16psi |
There's a thread here on the forum where a member did some temp drop testing and compared the STi and an eBay intercooler. Might have done a WRX one too. The eBay cooler was HORRIBLE.
I think I posted in it, ill have to look through my subscriptions. Found it: http://forums.nasioc.com/forums/show....php?t=2126473 Last edited by Black94Snake; 05-03-2012 at 11:12 PM. |
05-04-2012, 12:53 AM | #6 | |
Scooby Specialist
Member#: 290109
Join Date: Aug 2011
Chapter/Region:
SCIC
Location: Fresno, CA
Vehicle:2005 WRX PSM |
Quote:
|
|
05-04-2012, 08:46 AM | #7 |
Scooby Specialist
Member#: 159743
Join Date: Sep 2007
Location: Plymouth Ma
Vehicle:07 STI Limited TurboMike Tuned 2.35 |
And here i was thinking 5 people were going to recommend great reasons why my idea would work.
Your IAM is dropping from Rough Correction. normally means your running too much timing down low in the .40 - .80 range on your map. The Ebay Topmount will work just fine ive installed quite a few of them. |
05-04-2012, 03:56 PM | #8 | |
Scooby Specialist
Member#: 290109
Join Date: Aug 2011
Chapter/Region:
SCIC
Location: Fresno, CA
Vehicle:2005 WRX PSM |
Quote:
|
|
05-05-2012, 02:17 AM | #9 |
Scooby Guru
Member#: 140444
Join Date: Feb 2007
Chapter/Region:
NWIC
Location: Near Seattle, WA
Vehicle:05 Stage Free LGT ATP 3076, 6MT, AVO FMIC |
One thing bothers me about this approach... The ECU will periodically turn up timing until it hears knock, and then it will turn timing down. It will be doing this in a bunch of different RPM/Load cells, pretty much all the time, until you go back to a more conventional tune. How many knock events will the engine endure before you go back to a conventional tune? I have no idea how to predict that, or even guess.
Whereas if you tune it yourself you know have a pretty good understanding of just often the engine knocks because you see it in your logs when it happens, and you fix it when you find it. So you can minimize the amount of knock the engine has to put up with, and when you're done you have a tune that knocks rarely, if ever. You could perhaps use your idea to get an idea of how the timing table should be shaped, but you'd want to log the whole time to get an idea of how much knock you're getting. And if you're going to do that you might as well tune it yourself. |
05-05-2012, 07:47 AM | #10 |
Scooby Guru
Member#: 70395
Join Date: Sep 2004
Chapter/Region:
MAIC
Location: Da-boonies,Va
|
/\This. The knock system is a "reactive" type system and not a "proactive" type system.
|
05-05-2012, 10:44 AM | #11 |
Scooby Specialist
Member#: 159743
Join Date: Sep 2007
Location: Plymouth Ma
Vehicle:07 STI Limited TurboMike Tuned 2.35 |
yeah i already posted up things like this just didn't do it on nasioc. Ive been updating my thread in romraider's forum instead since i just get ebay TMIC answers here.
Ive been weary about trying it for that exact point... since i can adjust how often it adds or rechecks if there are knock events i would be putting the engine right on the cusp of detonation everywhere that fine learning is enabled. I thought about maybe taking away more timing per knock event and add less when it tries to add more to keep me away from that tipping point.. I could however see this used for a week timeframe. If you think about it on a car you want to get good mileage on you could run it for a week and look at the fine learning table, when setup right. Then make "true" adjustments to your base table the following week. Im going to try it on a wrx i just built with pistons and valves and an EGT sensor so i can produce real results without burning up stock parts. Thanks for the feedback though. |
05-05-2012, 02:26 PM | #12 |
Scooby Specialist
Member#: 290109
Join Date: Aug 2011
Chapter/Region:
SCIC
Location: Fresno, CA
Vehicle:2005 WRX PSM |
|
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Fine Learning Knock Correction | ScoobiePower | Open Source Reflashes | 12 | 03-16-2016 01:24 AM |
How come I Can't see feedback knock correction, fine learning knock correction? | AraSh | Open Source Reflashes | 23 | 01-08-2014 07:04 PM |
Something is wrong with my fine learning correction | manys | Engine Management & Tuning | 3 | 12-23-2010 10:26 PM |
Q about rough knock correction and IAM | scoobystas | Open Source Reflashes | 6 | 06-04-2010 07:31 PM |
IAM = 16, Knock Correction learning turns off. | jblaine | Engine Management & Tuning | 41 | 09-12-2004 10:49 AM |