CE Lancer ECU tuning

Good to see it’s working ok for you, your cruise is a little rich though... one of my issues is cruising at 14.3 is unacceptable! Haha.

I want solid 14.5-15 like stock. But it’s either mid 15s or 14.3ish. I’m wondering if the injector size calculation and MAF calculations are coded differently to the stock rom. I’m using values +\- what people say is the standard value and can’t hit the sweet spot and have a good idle.

Using the stock orange 210cc injectors and stock MAF.
 
I think the issue can possibly negated by using Evo MAF, and larger injectors.

But I wanted to build a proper base map you can flash onto a totally stock car :(
 
I'm the same, id love a stock-like cruise AFR, but its not easily done. I guess for reference, a mates car has been professionally tuned on a Link G4 and cruises at about 13.5, so im happy with how mine is, especially considering its an NA+T engine. I dont have fuel trims working, which is why im not able to really lean/richen it out further, this i think is an unrelated issue that im yet to really solve (not a biggie atm though).

All my research says you should be able to fine tune it regardless with the MAF scaling. My laptop died a while ago so i havent been able to fiddle with anything, but i run well enough to daily it at 14psi and not need to worry.
 
@donki
I remember you would always talk about getting the exhaust popping etc by richening up the mixture.
What exactly did you do to get that, ie, what rpm, what throttle position etc?
 
@donki

I remember you would always talk about getting the exhaust popping etc by richening up the mixture.

What exactly did you do to get that, ie, what rpm, what throttle position etc?

It's an odd situation that i need to preface: Exhaust popping is a lean (meaning leaner than stoich) exhaust gas condition. You would still have exhaust popping with a richer mixture, but its not as severe but can allow flames if its rich enough. As an example you would be able to pop and burble if you introduced an tiny exhaust leak prior to the muffler that would allow additional air into the system letting the mixture burn further/combust and causing popping - this is what antilag is (air from the turbo or intake manifold being fed into the exhaust stream).

In my situation I've simply put a kind of "If statement" in my tuning;
Code:
If RPM is >3000 (to avoid it occurring during any normal driving conditions)

and EngineLoad <30% (idle sits about 25-32% engine load, but we're above 3k rpm so load is only 0-5%)

then timing= -10 degress


It won't kick in at at any other time, even if i change gears casually at 4k rpm, it'll barely burble once during that rev drop into the next gear. I probably wouldn't recommend it. In the space of about 3-4 months i had a couple of mm of carbon buildup on my exhaust ports, which was less than ideal considering theyre already so small. Running E85 means that carbon builds up a LOT slower, but its still there. In addition, having the retarded timing set too badly cause mean you blow out chunks from your exhaust valves, which is a much bigger issue
 
Newer turbo cars retard timing by a fair bit when changing gears though in an attempt to get the combustion into the exhaust and keep the turbo spinning right?
 
Newer turbo cars retard timing by a fair bit when changing gears though in an attempt to get the combustion into the exhaust and keep the turbo spinning right?

Mostly correct. DSG/Auto Golfs, BMWs, Mercs, and a fair few others all do this, helps with the "sporty" note on a gear change, but they also introduce additional air into the exhaust system for the noise (and also for emissions purposes).
The difference in turbo spool up is next to nothing on a street driven car though, as like even with an Evo with the antilag setup from factory (for Mitsubishi to meet the WRC regulations) the spool up time difference was estimated to be less than 500rpm improvement with it modified.

Another point to note, a lot of modern turbo car (around 2012 onwards) run very stoich Air fuel ratios, hovering 14.7:1 for most of their runtime (emissions), falling down to around 12.5:1 under full WOT runs. This is more or less perfect power and emission extraction from petrol/oxygen mixtures. Direct port fuel injection, adjustable injector, valve timing and spark timing all mean the car's ECU's and sensors are already tuning and adjusting much more than what we could really do manually, and doing it bloody well. Modding any of these cars to get more power isnt easy, all that is usually done is upping the boost a little bit, because its been done so well already from factory/in the ECU on the fly.
 
Picked up a new laptop for tuning thanks to @sidewayz , so i can start putting all the theory ive read over the last like 2 months to the test.

I'm running ceddymods flashed to a stock 4g93 ECU without too many issues. Idle is up and down, but by over-riding it with the BISS it works well enough/I'm pretty sure my IACV unit is completely *fudge*ed anyway and has been for ages so that doesnt matter.

Just to confirm to myself/update everyone, my IACV is truely munted. Sometimes it'll work, and idle drops correctly during warmup (which is the easiest way to tell if its working properly through its whole motion). Its overridden by the BISS screw, which is the Base Idle Set Screw, which means its pretty much always set at 1000-1100rpm all the time, and if the IACV decides to get off its ass and work, then my idle will drop to about 750 +- 50rpm like normal.
 
Last edited:
As for flex fuel I believe the vr4 guys came up with something but you have to buy a special box from them.
From the code I have looked at and worked on I believe it is possible to use the subroutine that is used to interpolate between the stock ignition map and the warm up retard ignition map to set up flex fuel.
That whole sub is controlled by a table that uses a percentage vs temp so at say 80 deg it uses all of the ignition map and none of the ign retard map.
So instead of temp vs percentage you would use the flex input vs percentage between two of each ign/ fuel/ve and boost maps.

I do not currently have a car as I have no room with my work van but if somebody wants to try it I am more than happy to help out on the coding.

If you're still around, id be happy to test and try out any flex re-coding you can come up with.
I have a newer, pretty comprehensive xml for the evo5/6, based off Ceddymods, modified by Merlin, and prettied up by me, which may have a few extra lines of addresses that might be of help.
 
I posted in my own build thread, but to update here too, i've fixed a tonne of my issues by simple going to full time open loop. This mean the ECU isn't referring to the oxygen sensor to add or subtract fuel at all (which is how a car meets emission and such), but as the mapping is done well, theres no need to rely on the o2 sensor to provide the feedback anyway. The car in standard form will use the fuel maps in open loop mode, usually when above 40-60% engine load, or above a certain throttle position or RPM.
Full time open loop allows you to tune beyond the car's "restrictions" of the closed loop mode. I can force any Air Fuel ratio at any rpm, instead of having to deal with the automatic add/subtracting of fuel based on the o2 sensor feedback.

Edit: At the same time I also changed the MAF sensor stuff (dont recall the exact table name) where it reads HZ against Gm/s, and changed the gm/s in the 50-150 hz range so it was either open loop or this MAF setting but that helped fix my cruise
 
Last edited:
Posting moreso for my own tracking and information, feel free to ignore it may be somewhat shorthand

- Might need to retard idle ignition timing a bit for e85 maps. I have a theory that due to the ethanol having a faster flame front, the peak cylinder pressure occurs sooner after the spark event. At high rpm this isn't (as much) an issue because the combustion doesnt change at the time scale we're talking about (hundred of thousandths of a second for the flame front compared to hundredths of a second for crank rotation), but at low rpm the effect is more pronounced. I think my current timing is too advanced, causing "peak power" to happen at idle and causing the rpm to rise a bit
- above is when compared to the slower flame speed of petrol. My timing maps between e85 and petrol are the same in lower RPM/Load (not tuned yet), but the idle is a noticeable difference
 
yes e85 actually burns faster contrary to people's beliefs, "burns slower so you can add more timing".

if your not knock limited like at idle, low throttle opening or low boost etc then e85 will need less timing to make best torque/power, MBT.
 
Picked up a new laptop for tuning thanks to @sidewayz , so i can start putting all the theory ive read over the last like 2 months to the test.



Just to confirm to myself/update everyone, my IACV is truely munted. Sometimes it'll work, and idle drops correctly during warmup (which is the easiest way to tell if its working properly through its whole motion). Its overridden by the BISS screw, which is the Base Idle Set Screw, which means its pretty much always set at 1000-1100rpm all the time, and if the IACV decides to get off its ass and work, then my idle will drop to about 750 +- 50rpm like normal.

Was the iacv working normally before turbo swap/ecuflash? im seeing simmilar behavior but flashing back to stock map corrects it perfectly.

currently the best working setup sits at 850 warm idle, 1300ish on start up tapering down but mostly hovers around 900-1050 sometimes randomly. when it drops to the 850 its sweet. i believe this is the best we can get at this point, i was thinking difference in actual valve/throttle body could be a factor, but more likely some drastic changes in idle settings on the stock rom (not defined) vs evo rom.

id be interested to know what definitions you have and see if we could combine xmls etc, iv found a few and have settled on one now but i was switching back and forth to see how they varied, some tables are different, with things added and missing.

i think the ultimate answer to all the silly kinks is a 100% complete xml def file for both the ceddymod evo rom, and the stock 4g93. defined the same way, so *poo* can be exactly copied. iv spent MANY hours chasing the perfect stock idle and cruise even with stock injectors and maf there is some odd variances- revs dont drop how they would on stock map,and other small details, still good enough to drive and enjoy, reality is to some it wouldn't matter, but im super critical of these kinda things.
 
Yeah mate im still hanging around. What kind of stuff are you after.?
I have a 1.8 na maf h8 based rom with knock control that should be a direct flash to ce 1.8 maf ecu .
It can be modded to have larger maps until we max out the stock maf.
I think this would fix all the set up issues people have trying to run evo roms as it would run perfect out of boost until mapped.

If you're still around, id be happy to test and try out any flex re-coding you can come up with.
I have a newer, pretty comprehensive xml for the evo5/6, based off Ceddymods, modified by Merlin, and prettied up by me, which may have a few extra lines of addresses that might be of help.
 
So I thought I'd finally upload all of the files I use for my tuning.
It'll take a couple of days to find it all, but i'll get around to it. These are the files i've used to make a standard 86kW 4G93 into the 198kW/349nm 4G93+T, with the OEM ECU.

I know most of these files inside out, so if you have questions, please post them up and ill answer them. I guess i'll try to make a guide at some point on actually reflashing your stock ECU too.
Files will include Evo maps, CM5A maps, standard CE 4G93 AUDM and USDM maps, and whatever else i found useful/can find.

Edit: A few people may be a bit annoyed at the "release" of these files. I want to make it clear that these files are readily accessable online, on multiple websites, with a simple Google search. All files to the best of my knowledge, are free to distribute, as long as credit is given to the original creator. With this in mind, Ceddymods, Merlin, Tephra, evom.com forums, and several others have played a very important part in the creation of these files, and if you ever come across them on other forums, be sure to give them a thanks and a donation
 

Attachments

  • 25570101.xml
    66.8 KB · Views: 68
  • 24710004.xml
    18.3 KB · Views: 64
  • Lambda_vs_AFR.pdf
    19.3 KB · Views: 55
  • Cerom (1).XML
    12.2 KB · Views: 69
Last edited:
Back
Top