S13 ka24e injection time problem

[raddy]

Hi all. I have strange issue of injector timing at s13,ka24e, ecu 42FXX with turbo and 770cc inhectors. Problem is sudden injector timing increase at more or less exactly 4600rpm. After this rpm point inj.timing doesnt follow fuel map anymore and cause very rich AFR. Main fuel map, knock fuel map and upper gear fuel map are all same,even if related area was extremely leaned out,it doesnt change problem. But was quite surprise for me is that this problem is not present at 3rd gear,but is present at 4th gear and probably at 5th too,byt this wasnt tested. At 3rd gear inj.timing follows map all the way,just shift it up to 4th and after 4600rpm inj.timing jump up... It looks like 'upper gear"related issue,however cannot source it at nistune.
Thanhs for your ideas.

[Matt]

Could you please post the current tune file, as well as a log file showing the issue?

I cannot say off the top of my head what the issue is from the description, so we need the extra data to look at everything to see what is going on

[raddy]

Hi Matt,thanks for the reply. Attached are log files of separate 4th gear run and some 3rd and 4th gears runs at one session. Related maps are in test10.bin file. There is also test8.bin where I tested radical lean out at related are,but without any influence to AFR (after 4600rpm point).
Thank you for your help

[raddy]

rest of the files

[raddy]

Hi Matt,just by the chance,did you had time to check this issue?

[Matt]

Sorry I missed the update on this one. I've just gone through it now

What I suggest that you do is see how you go with adjusting the TP load scales, so change that from 38 max to 28 max

There is no increase with TP (load) so its not related to that. You have your O2 sensor disabled, so its not that (trims are near 0)

We get only the injection time jumping from about 10ms to 14ms

The other thing you can also try is lowering the closed loop scaler lower, so it goes from say 0-18 instead of 0-32 (where it is reaching almost at the moment)

[Matt]

attached is the changes I've made to test with. Let me know how that goes (with log file). If no good then I will setup on the bench

[raddy]

Hi Matt,thanks for reply. However,its probably small confusion here. Point of the log that you referring to is from 3rd gear run,which is completelly ok and afr is controlable by map,inj.time rise at 6000rpm is not issue. But when you check next run at this log,its 4th gear and at about 4600rpm is point of my issue...but only at 4th gear.not at 3rd....After this rpm point inj.time jumps from about 10ms to about 12ms....you can see difference if you compare area of about 4600rpm between 3rd and 4th gear run of this log and also in other attached logs with 4th gear runs. There is gear fuel map for this ecu used,by the chance,isnt its applicable for 4th and upper gears? If there is incorrect address for it,it can explain this issue.
Also,reason for relatively high scaling of load scale is that this is not final boost pressure,so there is room for more load. I just interrupted tuning when it was clear that is not possible to control injection time after 4600rpm at 4th gear...

[Matt]

But when you check next run at this log,its 4th gear and at about 4600rpm is point of my issue...but only at 4th gear.not at 3rd....After this rpm point inj.time jumps from about 10ms to about 12ms

Both times in that log it was when hitting 120km/h that we get the ~2ms jump in injection time. Fuel maps are the same, and normally gear maps are activated by rpm/speed ratios but I won't rule this out. Even the 37 and 39 log show the same speed (120km/h) where this is happening

I'll throw this on the bench and let you know

[Matt]

I can confirm this is 120km/h on the bench

It may be related to the gear fuel map, but I zero'd that out an it had no effect

Also I drop RPMs to 3300rpm and it still does it at the same speed (so not a gear coefficient)

Will need to check through the code this week and let you know

[raddy]

Thanks Matt!

[Matt]

Okay found the problem. There are actually 4 fuel maps

Now these ECUs (S13/U12 KA24E) are quite different from the others. There is no knock sensor, so there are no actual knock maps. However the ECU switches between the maps based on coolant temperature as well as engine speed (but not like a gear ratio, but fixed speed)

Use the updated address files and then use the 'copy all' button on the first fuel map to copy to the others

Unzip the address ZIP after downloading, and copy to Documents/Nistune/Address folder

2022-07-22 13_02_11-Nistune [U12(KA24E) (Version_ 1)].png

(170.32 KiB) Downloaded 391 times

So these are closed loop cut (112km/h + 4 km/h hysteresis) and what I've called O2 feedback temp and temp 2 (it isn't actually the proper closed loop STFT/LTFT switch temp, but I'm just reusing an address tag entry for this so we have it now)

[raddy]

Thanks a lot for your effort Matt,I will test it and post results.