1996 Mitsubishi Colt aka Mirage running lean and hesitation

Hi all im having problems with my car, Mitsu Colt GTI 4G93 DOCH engine 1.8L full engine rebuilt by me, under 1k miles since that.

Minor modifications:

4-2-1 headers, no-cat, Walbro high flow fuel pump, high comp pistons aprox 12:1 CR, 1mm oversized sleeves, innovate LC2 stand-alone WideBand.

Present real problems:

Low Power, Hesitation when engine is raced.

I have a Mitsubishi specific datalogger and this is what im seeing on the logs (i can provide any log if needed)

At idle:

-5% LTFT
-10%/-12% STFT
O2 readings seem right
No knocking
14,7 - 15,2 AFR

Under heavy load:

-5% LTFT
0% STFT under heavy load
O2 narrowband seems normal
1-4 Knock Sums
13,2 - 15,7 AFR (jumps a lot in that range)

Using 91 Octane fuel (best fuel i can get in my country)

Already checked for vacuum leaks just in case, I think im having the knock and the hesitation because a lean condition but what I dont understand is why the ECU is pulling fuel even when I would say engine is running leaner than it should...

Sorry for the long post. Any idea for what could be happening would be greatly apreciatted.

Was ecu tuned for the engine mods?

Sent from my SM-N920P using Tapatalk

Hi thanks for your reply and no, stock ECU without tunning, but i have 2 friends who have exactly the same pistons in the same engine and same make and model and year, none of them are experiencing any issues, we put the 3 cars in a rolling dyno, the other 2 cars won about 5hp and torque, mine lost 10hp and torque. and i connected my logger to their cars and there was no knocking nor lean condition. Also i suspected it could be an exhaust leak, then i ditched the old stock exhaust manifold and installed the current 4-2-1. issue persist.

Ok great to hear the other 2 run good with same build. I would start by testing compression on all 4 then do a cylinder leak down test on all 4 as well

Sent from my SM-N920P using Tapatalk

Ok I will do a comp test, i will need to borrow a tester on the weekend. Still i would like to think is not a mechanical related problem cause the engine is fully rebuilt and i think it was done with a lot of caution as was done by me and the other 2 friends who have the same build.

Also i cant see any oil or coolant consumption. rings are new, sleeves were bored out in a machine shop, valves, seats. seals the lot are new. Power balance test seemed ok too.

lets say you are in third gear cruising at 2000 rpm then you press the acelerator at 50% TPS and then the car just shakes a bit like for a second and the rpms doesnt move, then you got the kick and the car revs hard but with some stutering. same thing happens with the car at idle but less noticeable

Will do the suggested test, still in the meantime is there anything else that can make the car run lean and still make the fuel trims negative?

Thanks!

What is your air intake system ? Stock or cold air? Does this car have a mass air flow sensor? If so what are your readings at idle and wide open accelerating in gear? How about map readings at idle and wide open accelerating?

Sent from my SM-N920P using Tapatalk

Well first. I did a compression test, readings were about 155 - 140 - 140 - 150 psi. with a dry test but engine still warm.

NOTE: I already had to fix a bad valve seat where readings were like 60-130-120-130 even there the engine was running in 3 cylinders. I fixed this when the engine was rebuilt.

I dont know if i should worry about it but in all cyiinders the pressure was dropping very slowly, like at a rate of 10 psi in 30 seconds after the engine stopped cranking. The compression tester is a very cheap chinese thing.

Im attaching a pic of the sparkplugs I think they are ok, maybe a bit lean ish?

ibb.co/bG87yQ

And replying to your latest questions:

Intake system is stock with a MAF, however is a Karman Vortex sensor so as far as I understand, a dirty MAF is unlikely.

I didnt understand what did you mean by readings, I see that readings increase when throttle is applied, but I dont know what values are the correct, however Airflow seems to increase or decrease in conjuction with TPS sensor.

The car only has a MAF but not a MAP. Is a very old EFI system.

Additional info: As I am a weekend DYI Saturday i spent all day working on the car, Im planning on installing an Aftermarket (no chinese) adjutable fuel pressure regulator. Im working on an adaptation to install it and a fuel pressure gauge.

Fun fact: You know how much for a Gallon of fuel in my country? 9 dollars.

So it hurt a bit today when I filled in the tank, the car seems to run better, not perfect but hesitation reduced noticeably after the fill in... so Im still pursuing a fuel related problem, but i dont seem to pinpoint the issue.

Still i may be wrong in that case i keep listening to suggestion on where to look...

Can you get the maf readings either in voltage or grams a second for example your car should idle at around 2 gs and at wide open throttle you should be able to get over 100gs or over 4 volts, fuel injection relys very heavily on Maf readings . Your pressure dropping is a leaky gauge not a in cylinder issue.

Sent from my SM-N920P using Tapatalk

Ok that was a difficult one. The logger doesnt reads volts or g/s, it reads hz. So i checked the workshop manual and it stated the standard values for idling, 2000 rpm and racing. Im a bit confused by the results:

This is what the workshop manual says:

However Im not clear if the 2000 rpm test should be done at constant 2000 rpm or if I should race it till that speed and let off the gas and take measurement while speed is reducing.

I will probably do a constant 2000 RPM test, still im attaching one of the latest logs where i was letting the car idle and racing it. I can see airflow at idle matches specification, but at 2000 RPM is below the expected, it maybe because the throttle was closed at that time (TPS reads 12 something but is 0, this is common with this scanner, idle position switch is working fine)

I forgot to mention: the "*" refer to other specifications, *2 is the value for the DOHC engine so the value that involves my engine.

Hey that's ok we can convert for hz get a reading at steady throttle 2k and do a wide open throttle run while driving we want to know what the maximum air flow we can get, I know these Karmen vortex sensors usually aren't affected by dirt but sometimes they just don't read like they should

Sent from my SM-N920P using Tapatalk

Any update?

Sent from my SM-N920P using Tapatalk

Hello and sorry, it had been some busy and troublesome days. I just had one chance to drive the car and log it, i did a 10 min highway trip and did a WOT bit. I did it at about 50 mph in 5th gear and 2500 rpm. I pushed the pedal to the metal just for a few secs because what is my really ultimate concern is that the engine is showing some knock sums when load is applied and i was afraid of melting a piston or so.

I think this knocking comes from a lean condition that also is produced from the ecu pulling negative trims, hence this post about the negative trims.

Im attaching the log file, unfortunately is nothing fancy, dont show graphs, is not even OBD2 compliant, just raw data in columns but i tried to do my best to give some colours to the most important data. I think you can understand the whole file more than I do but the WOT bit starts at about row 1400.

What I observed is that even if it shows some little knocking sometimes, mostly when load is applied, when i am under WOT and the ecu is out of closed loop, the ecu is not pulling fuel anymore and no knocking occurs under WOT but it did under load AND in closed loop.

I hope you find this information usefull and thanks in advance for all the advise.

Hello all! Im refloating this topic because summer break is almost over and I still have the same problems...

I couldnt do much but I changed for all new injectors, also tested them, no leaking, FPR is working good also, no leaking. So im discarding this is a problem within the fuel system.

Now today im going to put a new OEM Spec NTK O2 Sensor, also I purchased some spray to clean the MAF.

Still I would like to hear more opinions.