NefMoto

Hello,

So  I have been messing up about with tuning my car after I successfully converted it to a t04e small frame from NA with the OG ecu - as you could read in the other thread - for about 2-3 weeks now.
So first of all CWMDAPP is set to 3 , which means pedal drives the Throttle plate angle using the map FPWDKAPP - right now throttle plate is about 3-7% higher than the pedal position - might change this later on cause of driveability issues for now @ idle its 3.13% AND timing is simple not torque based.

Trims - there are still no fuel trims , I'm unsure why, I didnt think cwmdapp would disable fuel trims but it seems like it does because before OG flash read it had trims. Some people suggested I should check out NORLA value but I am unable to do because it is not defined yet, and finding single bits is quite hard for me still even if I know the "Region" it resides in - some help would be appreciated :)

Fueling and Idle - So since I got no trims I am adjusting fuel based on lambda control in closed loop obviously.
 since cwmdapp there is no idle control so my idle RPM is around 1000-1300 flucutating cause of lambda control and rich condition , if I disable o2 regulation it doesnt. So today I changed FKKVS, after reading some more because originally in the FR it says the axis is tevfa_w , which you will see in logs ranges from 0-0.68 ms in my case, but the FKKVS map ms axis goes up till 12ms on this NA soft, so I though it must be the calculated injector on time. - adjusted FKKVS at the low ms injection time and the rpm I was seeing and you will see what happens :P vs the non adjusted map ( last link of the logs)

Limp - so originally I would get epc light at part throttle real fast cause of torque monitoring( because it is still active regardless of cwmdapp ) , so I have adjusted my requested load ( which caps out at 100% sadly it seems) and torque requests and KFPED ( and the load/rpm in KFMIRL) BUT not KFMIOP . This seems to have fixed most of these issue, but however today I had it again , as you will see one of the logs -  gonna have to adjust that still. Still have to match some areas in requested/actual load but for the most cause they are "aight" - I think  :D

Any tips , help etc is much appreciated as long as its constructive :P Thank you if you read it all. There is still much to do.

Idle log after FKKVS adjustment - https://drive.google.com/open?id=1MQARslVrlNf1HSV8M-NibmzUqwdLHWwO

Some pulls log and limp mode(above) at 50 second markish ( I think) - https://drive.google.com/open?id=1GSoUY_mTfZ5ZWAtZupLGF3PWoo_OocHX

First start of the day ( without FKKVS adjustment ) and light driving until pull location - https://drive.google.com/open?id=1Uubz3doYqq-dSnyJIRw_aSQBmMNR16ke

Anything ? Also had to upload logs on Drive since for some reason it does not let me upload them as attachments  ???

The Torque monitor limit 2 thing seems to happen at a very specifc pedal pos / torque request and actual load/requested load so I havent gotten it recently. However after 9 versions I am gonna branch off here.

The first one I'll leave it like this KFPED values mostly stock, torque requests and requested load is matched to actual load - however I am gonna try something different.

For the second one I am going to use the ME7 tuner wizard to generate a KFMIRL map , the corresponding IOP and disable torque monitor 2. And match the KFPED / torque requests to the requested load -> actual load if that makes sense - even though load caps out at 100% this way the torque request will correspond to the actaul load and I can adjust AFR with the LAMFA ( since again NA doesnt have bts fueling) , instead of adjusting the requests from Kfped so I can adjust afr from lamfa.

However I am still having issues with idle. It idles at quite high rpm so Ill try finding idle RPM map and change but I dont think it has any effect at all. I would want to lower RPM to about 1k max. Currently at 0% pedal pos the throttle plate angle is 3.13% , I could reduce this to lower the RPM BUT I am afraid it will give issues when the car is in neutral  or on gear shifts.
Not to mention on warmup lambda control is interfering so RPM is fluctuating since on warmup its enriched so its trying to correct it but it cant.

Also some minor issue is that sometimes the car is buckling when in gear deacceleration from 2000 to 1500 RPM.

RLVMXN

RLVSMXN

Hm I found KFMIZUOF yesterday , and setting it to 100% on all points should disable torque monitor 2 + the 2 temp scalars  from I've read on this forum and s4 wiki, should I also find these maps and change them ? Only torque monitor 2 is giving issues. Ill try finding them.

Btw how do you find single bits properly ? Even If I find them I am not sure if they are right, is there any way to confirm this, changing them randomly is scary because ususally there are multiple single bits next to eachother.

1 change at a time upload your original file and I'll see what I can find. Removing load caps and mxn smxn should get you boosting without limp

http://nefariousmotorsports.com/forum/index.php?topic=15390

here is the original bin.

But thats the thing. There is no actual load cap on NA, my actual load correctly goes above 100% , fueling follows it and no limp in boost , limp only happens in a very certain pedal position, ( about 20-30% pedal post and about 2.5k rpm - I assume because of torque caps so Ill remove those). However something caps requested load at 100% - somebody said its hard capped at compile ,even though torque request can go above 100%( so I can use lamfa to fuel properly) but not requested load, if I could make it go above 100% I could match it to actual load , just so it looks nice on logs :D.

Also for fuel trims I was looking for NORLA but its hard for me that find that cause its a single value thats why I asked :)

load caps are at 4D*** AND 4D***. Set both to matching same offset factor as irl  ;)

wow thanks!
heh I ve seen that you edited post. I checked out those adresses and they gave me 3.984 with kfmirl's factor? so I assume they are not right , unless I did something wrong.

should be right addresses need to look at em in 16bit low high

Nice, I changed them to 16bit low high and they are showing 99.987 which is true. Thank you so much.

So I am safe to set this to the actual load I am getting max like 143 ?

No bueno, I changed them to 140 , yet my most current log still shows that requested load gets capped at 99.984.( Actual is fine like I said before)

I'll check again when I get back to my computer

Sure.

Meanwhile here is the latest log: https://drive.google.com/open?id=1-DjGYC9Is2xCKs-cenKYQn4I0iczJAfC

Ive adjusted KFPED/torque requests even more to match the load , gonna flash that tommorow.

So today actually I forgot to flash the "fixed" file and got limp again at 2.5k rpm'ish :D Its fine you just have to cylce the key. Hopefully I wont forget to flash it tommorow.

example: (https://i.imgur.com/7aZnvpR.png)



So a couple of other things. My cluster reads that either I am "eating" gas way too fast or idk but the gauge is dropping very fast. I assume this could be related to KVB which I found ( hopefully right 1A00C - I would appreciate if someone could verify) , so I am going to try to change it to my actual injectors ~370cc. as the stock value is about 185 cm3/s , when I flash a file , the gauge actually goes up but after start and a bit of driving it drops so thats why I am guessing KVB has an effect.

I still cant properly find NOLRA value and its bugging me cause no fuel trims :( anyway  I am thinking using FKKVS fixer to fix lambda control.

Also probably gonna turn torque based ignition back to get a bit more control that way over idle and lower 0% pedal position throttle plate angle to about 2.5 % from 3.13% because it seems like its too much.
Which is probably also related to KISRM. I was thinking and probably should change this because I have switched to the 1.8t intake obviously, which is probably smaller than the NA one with the changeover too - also on changeover the KISRM value changes according to the FR , so probably should also take a look at that.

What do you think ?

Allright. I seem to have fixed all the torque intervention issues and things like that. No limp mode from 1.2k RPM to 4.5k RPM in any pedal position I'm quite sure.

I just made a log today to drop it into Guitar's FKKVS fixer to fix the stfts ( = lambda control - still no NORLA has been found :( ) and hopefully they will be near perfect.

However a bigger issue right now is cold start.... I can basically not start up when engine is cold - well I can if i crank for like 5-7 seconds or I press pedal to let more air in ( and more fuel...) , and if it starts up it runs very very bad, since its expecting way more air than 3% throttle plate angle vacuum , and just basically wasting fuel.. after the cat heating sequence turns off the car literally runs like it should apart from idleing at 1.2k RPM.
For that I am planning to lower 0% pedal position to 2.5% throttle plate , and hopefully that will lower it to 800-1000 rpmish which would be decent. One issue I might run into is lights+ ac at idle, right now they already make the RPM drop a tiny bit, since I am always letting the same air in and same timing, so for that I'll turn back CWMDAPP to 2 so I can use proper ignition angles to I can use torque reserve intervention etc. to help out a tiny bit.
Also I am having a hard time finding these maps to turn off cat heating. byte values are nono( CWKONABG or FKHAxx), so I am trying to find the maps maybe but yeah..

If you really want to do turbo in N/A engine dont change CWMDAPP because this generate a lot of problems for normal driving. You can do 2 hacks but remember that ecu will be seeing wrong data and could work somethimes unexpected. Modify HFM MAF linearization to lower load, where turbo boosting and add fuel in FKKVS table to be equal to air what ecu dont see. For thottle operation, edit KFWDKMSN table. If you something do wrong you can damage engine. You must checking all time on AFR gaugage.

The motor is already converted there is no going back. And not really I dont need to do hacks apart from cwmdapp. The MAF has not been changed since  its a different version than 1.8ts also 3 inch/2.75 inch, can read about 360 g/s or so. KFKHFM was also all 1's in stock file. The ecu can properly see actual load going above 100%, also ps_w correctly sees 1.5 bars ( 0.5bar from turbo). I already modified LAMFA and torque requests so I get richer mixture in boost areas, and funnily enough the NB also can detect if its "high rich" or low rich - for example when I start richening mixture ~14 afr it reads about 0.7 volts when Its down to 13~ afr requested it sees about 0.88v :) Lambda control takes care of part throttle fuel, that is why I am fixing FKKVS so it doesnt need to fix much.

KFWDKMSN has no effect in dbc mode, everything is driven by FPWDKAPP.

So I just spent 3 hours today trying to find the bloody CWKONABG... I have tried every 1 value in the PROKON area, and it seems like on the very last one aftee 3 hours and like 9 writes/changing one by one I managed to find it , but I am not 100% still. Could somebody confirm if - 18198 - is truly right. I have logged b_kh and it was 0 when I started the car up , on every other try it was always 1.

Gotta confirm tommorow though when I start it up.

Nope.. b_Kh still active when I start up... I went thru all the obvious prokon area , if its lower down in the adresses I dont know how to find it :/ Guys I would really appreciate this if someone could help me out , I am basically dumping raw fuel in exhaust like 12-13 AFR for a good 3-4 min on idle, just wasting fuel.

Well I couldnt find CWKONABG but I found FKHABMN and the individual maps in BBKHZ so I disabled those.