OFT - Guide To Dialling In Your OTS Base Tune

[steve99]

Originally Posted by KoolBRZ

@Wayno, I'm trying to adjust my Base Timing B table to reduce knock, and I noticed something. When I first start logging my tune the knock is higher, and reduces the IAM more, than after I've brought IAM back to 1 and started pushing it again the knock isn't as much and the IAM also doesn't go down so low. This makes me feel that my car is learning and mapping knock on some kind of table in the ECU, and if I could only download that table I'd have the perfect map of where knock is happening, without having to log, read, re-tune, and log again. Do you think it is possible to download this table with a tactrix cable or something.

their is a FLKC (fine learning knock correction table ie learned knock table, so yes ecu learns where knock happens and stores it. While we log FLKC and see it instantaneously its also stored for a period in a table similar to the one below.

their is a whole lot of algorithms applied by ecu to work out if it will decrement or increment iam and when it will return the flkc timing it stores. When you see positive FLKC this is the ecu returning some stored flkc

that learned table is storred in ram and gets cleared on ecu reset or flash

their is a whole lot of info on romraider site

see the flashing-RESET-Learning link below

the romraider guys have done it with learning view but its not set up for brz/86 ecu

[ztan]

Originally Posted by KoolBRZ

@Wayno, I'm trying to adjust my Base Timing B table to reduce knock, and I noticed something. When I first start logging my tune the knock is higher, and reduces the IAM more, than after I've brought IAM back to 1 and started pushing it again the knock isn't as much and the IAM also doesn't go down so low. This makes me feel that my car is learning and mapping knock on some kind of table in the ECU, and if I could only download that table I'd have the perfect map of where knock is happening, without having to log, read, re-tune, and log again. Do you think it is possible to download this table with a tactrix cable or something.

Some of the reduced knock perceived by the ECU is the knock sensor processing learning some of the background valvetrain/mechanical noise. Highly difficult to unravel this code and I'm sure that it would be quite easy to tune the knock sensor catastrophically. Having said that, if you want to look at bits of the code, I have some disassembled.

Otherwise:
What is your ROM ID?
FLKC is stored in RAM and referenced by an offset value - I can try to get the addresses for you, but you'll have to pull 25 extra values each logging cycle to get the table values for a full learning view as per Steve's post. I log FLKC and offset cell as well as ROM and load and get all my logs processed in a single hit using MATLAB.
My current log speed is 25Hz pulling 35 parameters.

[KoolBRZ]

Originally Posted by ztan

Some of the reduced knock perceived by the ECU is the knock sensor processing learning some of the background valvetrain/mechanical noise. Highly difficult to unravel this code and I'm sure that it would be quite easy to tune the knock sensor catastrophically. Having said that, if you want to look at bits of the code, I have some disassembled.

Otherwise:
What is your ROM ID?
FLKC is stored in RAM and referenced by an offset value - I can try to get the addresses for you, but you'll have to pull 25 extra values each logging cycle to get the table values for a full learning view as per Steve's post. I log FLKC and offset cell as well as ROM and load and get all my logs processed in a single hit using MATLAB.
My current log speed is 25Hz pulling 35 parameters.

My ROM ID is A00D. Are you using a Tactrix cable or equivalent to log this? I'm trying to decide if I should get one to make adjustment of timing easier. I have a phnatom ESC, Tomei UEL, and custom AVCS mapping and I'm getting lots of knock on the OTS OFT ESC tunes. This is while running 92 Oct E10 winter blend, which doesn't knock as much as the E0 summer blend.

[Wayno]

I don't think there's any advantage to tactrix over oft for adjusting timing. With the sheer amount of data you need to connect over a long period of time, the extra resolution of tactrix won't add very much.

[JB86'd]

I've been running with the Di/PI ratios, and load limits.It's definitely better than the original OTS map, but the AFR curve is a bit lumpy from 3000-4000 still and it's still rich from 4k to redline.
My CL LTFT isn't bad, so does it make sense to revert to OTS load limits, scale only OL maf, and then once that looks good, log and mess with load limits to smooth that 3-4k area? Last time I had load limits and flashed a lean version MAF scale it was too lean in a lot of places.

Latest 3rd gear pull for reference:

http://datazap.me/u/joshbustos86/jdl...4&zoom=112-266
@Wayno @steve99

[steve99]

Originally Posted by JB86'd

I've been running with the Di/PI ratios, and load limits.It's definitely better than the original OTS map, but the AFR curve is a bit lumpy from 3000-4000 still and it's still rich from 4k to redline.
My CL LTFT isn't bad, so does it make sense to revert to OTS load limits, scale only OL maf, and then once that looks good, log and mess with load limits to smooth that 3-4k area? Last time I had load limits and flashed a lean version MAF scale it was too lean in a lot of places.

Latest 3rd gear pull for reference:

http://datazap.me/u/joshbustos86/jdl...4&zoom=112-266
@Wayno @steve99

waynos maf scales work most times but some cars benifit from a bit of custom maf scaling look like yours might be one of those. Theirs nothing rearly wrong with it at present but it is running a bit richer than commanded afr at highr rpm

[Wayno]

Originally Posted by JB86'd

I've been running with the Di/PI ratios, and load limits.It's definitely better than the original OTS map, but the AFR curve is a bit lumpy from 3000-4000 still and it's still rich from 4k to redline.
My CL LTFT isn't bad, so does it make sense to revert to OTS load limits, scale only OL maf, and then once that looks good, log and mess with load limits to smooth that 3-4k area? Last time I had load limits and flashed a lean version MAF scale it was too lean in a lot of places.

Latest 3rd gear pull for reference:

http://datazap.me/u/joshbustos86/jdl...4&zoom=112-266
@Wayno @steve99

You just need to keep everything the same and change the last few values of the maf scale to these and you'll be done.

64.60
85.50
110.20
129.00
145.50
175.00
203.50
244.00
301.00

Also make sure you have load limit at 3700/3800 is 1.1 or something like that and not 1.0.

[JB86'd]

Originally Posted by Wayno

You just need to keep everything the same and change the last few values of the maf scale to these and you'll be done.

64.60
85.50
110.20
129.00
145.50
175.00
203.50
244.00
301.00

Also make sure you have load limit at 3700/3800 is 1.1 or something like that and not 1.0.

Sweet, will do. I'll post a log once I put some miles on it. Many thanks.

[Wayno]

Originally Posted by oldgunfan

Ok I got about 500 miles on my last flash, ver2.7 stage2 UEL 93oct tune with Wayno's load limit and fan mod. I just did a 3rd gear pull it looks fat like 11.7 to 12.3 afr. any help.http://datazap.me/u/oldgunfan/log-14...0&data=1-2-3-9

It looks like your load limit is too low at 3800 which makes it go too lean. Make it 1.10 instead of 1.00.
MAF scale looks fine so don't touch that.

The 1.00 load limits are only for the Tomei/PSR/Gruppe-S headers, not the ones with better flow between 3k-4k like JDL/OFT/Revworks.
The example map thread has the best examples to copy.

[oldgunfan]

Originally Posted by Wayno

It looks like your load limit is too low at 3800 which makes it go too lean. Make it 1.10 instead of 1.00.
MAF scale looks fine so don't touch that.

The 1.00 load limits are only for the Tomei/PSR/Gruppe-S headers, not the ones with better flow between 3k-4k like JDL/OFT/Revworks.
The example map thread has the best examples to copy.

Thanks for the reply, I did a little playing around with the MAF scale after the last pull see what you think now. http://datazap.me/u/oldgunfan/last-l...ata=1-2-5-9-14

[oldgunfan]

Sorry my bad that was the leanest I went, I have it richer now. I'll record a pull tomorrow and post it.

[oldgunfan]

Originally Posted by Wayno

It looks like your load limit is too low at 3800 which makes it go too lean. Make it 1.10 instead of 1.00.
MAF scale looks fine so don't touch that.

The 1.00 load limits are only for the Tomei/PSR/Gruppe-S headers, not the ones with better flow between 3k-4k like JDL/OFT/Revworks.
The example map thread has the best examples to copy.

Ok I made another pull this one was a good 3rd to 4th gear pull on a 9% grade. It runs like a raped ape, But it looks like it starts getting to rich over 5000 rpm. BTW I looked and you were right, my load limit 3800 is at 1.00. do you think I should still change it?

[oldgunfan]

Link didn't work???? http://datazap.me/u/oldgunfan/dialed...1-2-3-4-5-9-14

[power_of_ra]

Hey everyone,

total newb here trying to quiet my cold start rev a bit so i took my stage 2 UEL 91 tune and changed the post ignition timing start values to -6.99 from -15. saved it as a new .bin and compared - looked all good.

when i tried to upload to my OFT, i get this error [see attached image].

B00C cal ID using B00C .bin and .xml from OTS 2.074. What am i doing wrong?