Road tuning with OFT

[Kodename47]

Originally Posted by steve99

one thing they did not mention in the video above is that running richer generally buys you knock resistance so you can run more advance and make more power, there are many interactions when tuning its not simply add more timing make more power everything it a tradeoff and has other effects some undesirable.

Very true, but it does show that within limits that AFR swings doesn't have a big impact on power. What you have to be aware of is the resultant exhaust gas temperatures

[BuBlake]

Originally Posted by Kodename47

The 2 maps you're on about are AVCS active, usually with the higher values, and AVCS off. The advance map will likely be called ignition advance or maybe advance A. There should only be 3 ignition timing maps that look correct.

Okay, so the values I changed were likely in AVCS on(base table B). Now the question is when does AVCS turn on? Also assuming AVCS stands for active valve control system.(google)

knock correction advance max A is probably what you would call the advance map. It contains values between 0 and 6 degrees and has load and rpm parameters on a large graph just like base timing A and B...

From what I understand, the ecu uses values in the active cell from knock correction max advance A, then multiplies them by IAM(which is 1 because I have no knock), then as far as I know, the only thing it does with that value is display it as KC learned in datalogs. lol... I'll go read some more, please don't kill me.

[steve99]

Originally Posted by BuBlake

Okay, so the values I changed were likely in AVCS on(base table B). Now the question is when does AVCS turn on? Also assuming AVCS stands for active valve control system.(google)

knock correction advance max A is probably what you would call the advance map. It contains values between 0 and 6 degrees and has load and rpm parameters on a large graph just like base timing A and B...

From what I understand, the ecu uses values in the active cell from knock correction max advance A, then multiplies them by IAM(which is 1 because I have no knock), then as far as I know, the only thing it does with that value is display it as KC learned in datalogs. lol... I'll go read some more, please don't kill me.

Have a read here

http://www.ft86club.com/forums/showthread.php?t=61820


Base_TIMING_B is used during normal operation
Base_Timing_A is for limp home mode

[Kodename47]

Originally Posted by steve99

Base_TIMING_B is used during normal operation
Base_Timing_A is for limp home mode

Kinda, the AVCS (VVT) isn't active on cold engine start and doesn't become active immediately. I assume that's why that map is there as changes in overlap will affect the dynamic compression ratio.

[BuBlake]

Originally Posted by steve99

Have a read here

http://www.ft86club.com/forums/showthread.php?t=61820


Base_TIMING_B is used during normal operation
Base_Timing_A is for limp home mode

Great write-up. My question is now, when logging, if I have no "knock correction"(all zeros), does that mean in all cases that the ignition timing actually occurring is Base Timing B added to the knock control advance max A table?

Also, to which table should I add advance? Base Timing B or the KCA table?

[Kodename47]

Originally Posted by BuBlake

Great write-up. My question is now, when logging, if I have no "knock correction"(all zeros), does that mean in all cases that the ignition timing actually occurring is Base Timing B added to the knock control advance max A table?

Also, to which table should I add advance? Base Timing B or the KCA table?

"Knock correction" on the logs is FBKC, "Ignition Correction - Fine" on the logs are FLKC. Both are equally indicative of knock. If both are 0 then you are constantly getting 100% of your ignition timing total.

As to which table you add to is up to you. Is this on E85 or petrol?

[steve99]

Originally Posted by Kodename47

"Knock correction" on the logs is FBKC, "Ignition Correction - Fine" on the logs are FLKC. Both are equally indicative of knock. If both are 0 then you are constantly getting 100% of your ignition timing total.

As to which table you add to is up to you. Is this on E85 or petrol?

As kodename47 says adding timing to either base b or kca max A will have same effect overall timing wise. However the KCA max A value is what the ecu has at its disposal to correct any knock that may occurr so if you take away from this area it limits the ecus ability to correct knock. This isalso the alue that is multiplied by the IAM for rough knocck correction so adding or subtracting amounts from KCA table will effect the amount of retard applied if your IAM drops.

once the eu has used up the values in kca table it appears it can no longer reduce timing via iam fbkc or flkc . I suspect that the values in the kca table are selected to be larger in the knok prone areas such that when the iam applies a percentage reduction more timing is pulled in certian areas ie where kca value is larger.

[Kodename47]

Originally Posted by steve99

once the eu has used up the values in kca table it appears it can no longer reduce timing via iam fbkc or flkc . I suspect that the values in the kca table are selected to be larger in the knok prone areas such that when the iam applies a percentage reduction more timing is pulled in certian areas ie where kca value is larger.

Are you suggesting that if the advance table was all 0 then the ECU would stop correcting knock? I know that the IAM becomes fixed if the table value is below a set range, but I've always assumed that FBKC is always active. It makes sense that FLKC cannot occur if the tables are at 0, but then would the ECU not use FBKC in its place?

Personally, if you have a good base to start on, I like to add to the advance table.

[Kodename47]

I found the answer

Originally Posted by merchgod (Romraider)

Yes, both feedback correction and fine correction learning are still active when IAM = 0 due to knock. And will correct to their respective limits. The only difference is that when the IAM is 0 or 1 and then after switching to fine correction mode, it will not enter rough correction mode unless the current fine correction(applied) value is positive (along with the normal requirements for switching modes).

[steve99]

Originally Posted by Kodename47

Are you suggesting that if the advance table was all 0 then the ECU would stop correcting knock? I know that the IAM becomes fixed if the table value is below a set range, but I've always assumed that FBKC is always active. It makes sense that FLKC cannot occur if the tables are at 0, but then would the ECU not use FBKC in its place?

Personally, if you have a good base to start on, I like to add to the advance table.

have not actually tested that out with the fbkc but that knock control method only seems active in lower to mid rpm ranges have not observed that mode of knock correction above about 4500, so you may still be limiting any knock control at higher rpm if you dont leave suffient headroom in the kca table , not something i rearly want to test , adding to advance table should be fine as it give the ecu more to play with especially on iam side, i just dont feel comfortable removing to much from the knock correction advance table, just prefer to remove it from base timing b, then i know i am not hindering the ecu if it keeds to correct knock due some bad fuel or similar.

[steve99]

Originally Posted by Kodename47

I found the answer

hopefully that applies to brz/86 ecu as i think merchgod left before the advent of our ecu, i would reckon it would run a similar method.

I think someone quoted 2.9 as the value in the kca table that is the cutoff for rough correction, i figure the kca table has that strange curve or values for some reason probably known to the subaru/toyota engineers so i like to leave it alone and make adjustments with base timing B , that just my preference based on my liited knoledge.

[steve99]

@Kodename47

dug up an old log where I had a bit of knock.

It using Tactrix to log and parameters provided by open source romraider guys but it appears to show that KCA can be 0 or negitive ? and that even when KCA =0 the FLKC correction is applied to the total timing value.

Assuming the log parameters RAM addresses and scaling are correct.

although in some areas it does not seem to add up unless some other compensations are being applied.

http://datazap.me/u/steve99/98-knock...zoom=2020-2314

[Kodename47]

I just read the whole knock control thread on Romraider again, it is as I thought. FBKC and FLKC are always active so long as you're within the defined operating ranges and can apply up to their independent set maximum retard value regardless of IAM and the advance table value.

I need to look deeper into the appearance of "Ignition Correction - Fine" on logs as it's hard to determine on that thread whether it shows as immediate correction as per FBKC or just as historic knock, or pre-learnt knock if you will. I've gone back and forth on what I think happens, if it's the latter it's much harder to locate the exact area that has an issue.

[solidONE]

Originally Posted by steve99

@Kodename47

dug up an old log where I had a bit of knock.

It using Tactrix to log and parameters provided by open source romraider guys but it appears to show that KCA can be 0 or negitive ? and that even when KCA =0 the FLKC correction is applied to the total timing value.

Assuming the log parameters RAM addresses and scaling are correct.

although in some areas it does not seem to add up unless some other compensations are being applied.

http://datazap.me/u/steve99/98-knock...zoom=2020-2314

That log was done with tactrix? I might get one just for the logging capabilities. Would save me a lot of time.