My car is just crazy, bad ECU?

Okay, I will do it and I will test it xD

 

By the way, a few months before discovering this problem, I rebuilt myself the K04 with a good brand CHRA (Melett), but checking last WOT logs, I can see turbo is suffering so much to hold 15/16 psi, very high WGDC over 4,5k RPM.

Maybe I have not sealed properly the intake housing of the turbo? or something else?

 

Maybe you shouldn't rebuild the turbo. Is it balanced? Was there even anything wrong with it or did you rebuild it just to rebuild it?
It amazes me that you've cheaped out so many times to repair "issues" and have only managed to shoot yourself in the foot each time.

 

I rebuilt because that.

 

Upping your oil weight likely would have fixed that. There are other causes as well; most of which are documented on this site.

 

Yeah I know some people changes to 5w40, and Mazda USA said the problem is the crankcase ventilation and they replaced valve covers and inlet pipes etc. But the truth, talking about why I rebuilt the turbo is not going to help me atm.

I trust Mazda when in the manual says to use 5w30, and Mazda Spain wants over 1600 dollars for the gen2 valve cover.
I bought the full CHRA balanced, just had to mount in the housings.

 

Smoking turbos are not turbo problems they are crankcase pressure problems

 

Yeah I know.

I will repeat my last question to see what people thinks. A few months before discovering this problem, I rebuilt myself the K04 with a good brand CHRA (Melett), but checking last WOT logs, I can see turbo is having a bad time to hold 15/16 psi, very high WGDC over 4,5k RPM.

Maybe I have not sealed properly the intake housing of the turbo? or something else?

 

You seem to be going all over the place with your issues and diagnostics. Tackle one thing at a time. The stock K04 will not hold over 17-18 psi max towards redline. It just doesn't have the strength to do it since it is such a small turbo. When you "rebuilt" the turbo did you just order a new preassembled CHRA and installed it in the turbine/compressor housings? Or did you disassemble the internals and try swapping the stuff out yourself?

Turbos have to be balanced properly. Your engine spins upwards of 7k rpms while the turbo itself is well over 100k rpms give or take.

Like EB said, leaking turbos can be crankcase pressure issues. They can also be issues with oil feed lines etc if the turbo requires a restrictor to prevent too much oil from flowing through it. Why not start your diagnostics all over again because so far this thread has been hard to follow. What is the main issue? The smoking? If so I said to remove the downpipe from the turbo and inspect the inside of the turbine housing for oil residue. Should be plenty obvious and easy to do.

Next if there is oil there then begin to check other things that affect crankcase pressure. IE your PCV valve, vacuum leaks, plugged catalytic converter etc.

 

If you just read the first post, you will know what's the problem and all the updates or changes that have happened in my car since I discovered the issue. You don't event have to see datazap logs, just read the text.

It's not the smoking turbo, car is not smoking anymore since I rebuilt the turbo.

I bought one of the best European brands of preassembled and balanced CHRA and installed it.

 

So what you're worried about is your LTFT being excessive? Within +/- 10% is usually normal and in range of what ECU's like to see before setting a code either way.

As for cleaning your ECU connections and then the issue seemed ok, did you add dielectric grease to the connectors? Terminal fretting/corrosion builds due to heat build up and friction between the surface of pins. By adding the grease you keep the pins cool and prevent moisture from building on their surface.

If your LTFT was around -25% then you probably need to change your oil. When fuel gets dumped into the crankcase because of a rich condition it thins the oil and allows it to be burned easier since the rings won't scrape and pull the film down. This is why we have PCV systems. When the system is in vacuum it pulls the rings against the cylinder walls to allow for a proper seal. Any weakness in the PCV system is preventing the rings from doing their job.

 

I am not having LTFT problems since 2 weeks.

Problems are different deppending of the car wants, thats why I said in the tittle "My car is just crazy"

Summary:
1. First all the parameters of the ECU with my custom map were bad (load, ignition advance, boost, wgdc, afr, desired afr, etc etc)
2. Go stock and flash "stock map", record some logs and everything looks OK except the desired afr, the afr idling that oscilates so much and big negative LTFT
3. After that, clean and reconnect First O2 Sensor and ECU, and desired afr and afr idling were perfect for 1 or 2 days, then going bad again, and still big negative LTFT.
4. Then appear 2 CEL, Rich at idle and High voltage secondary O2 sensor, I put a new Denso O2 sensor and both error disappear, reset fuel trims but everything still same, rich idle, rich afr desired but apparently LTFT stays under -6%
5. I make some logs but now AFR WOT is lean and everyday the desired idle afr and the idle afr is getting richer (I am running same map as step 2 and in step 2 AFR WOT was good)

The thing is, 9 months ago I rebuilt the turbo myself with a prebalanced good brand CHRA, I used the car for 5 months without making a log to check everything was OK (soft use always, no wot, no hard drive), then one day I made a log and discovered everything was working bad. Before the turbo rebuild everything was working good.

I don't know if reconnecting ECU fix something or just reset something. I just try different things to collect info and car reaction and luckily someone appears who may know what is wrong with my car.

To know every step, every change, every test, it's needed read all the posts/comments.

 

So you aren't running a tune for your mods?

 

When you disconnect the ECU or the battery the ECU resets itself. It learns the fuel trims, throttle body position, accelerator pedal position etc. All those values are lost. So initially you're seeing normal fuel trims because the ECU is starting from scratch and calibrating. After a few drives cycles or a hundred or two miles it goes back to how it was before the reset.

 

I was using a custom map when car started failing, but now I am running stock intake, stock testpipe, stock exhaust, only modified is downpipe is catless, but I am running a Versatuner OTS map, calibrated for stock airbox intake, and allows to use catless dp.

Are you sure? Yesterday when I disconnect and connect the ECU again, desired afr and idling afr changed from 14.59 to 14.70-14.78 but fuel trims remained at the same value it was before reconnecting it.

 

I'm out; just gonna keep going in circles.

 

Your ideas:

1- Check MAF connector/wirings >>>>> Done
2- Unplug Secondary O2 Sensor >>>>> Next Saturday I'll

Can you understand I can only work on the car on Saturdays, that's why I didn't unplugged the secondary o2 sensor yet.

Making tests is going in circles, okay. Desired AFR doesn't matter (Doesn't matter if it's 10:1 Desired AFR idling), okay.

 

Does your car feel like it is running ok though or does it feel rough? I would post an idle log and possibly a wot pull even if it isn't a full one.

As far as secondary o2 I would say if you're a gen 1 speed 3 I don't think that even matter and can be unhooked and the bung plugged. If you only "cleaned" the primary o2 sensor I would suggest replacing it instead as that doesn't always work.

Also, please make a simplified list of current problems. With these three or so things done we can probably better help and understand your issues

 

Thanks for answering. When I had the LTFT in -20/-25%, car has some hesitation/jerks/pulls in low rpms, now LTFT is under -6% and car feels normal in low rpms, no hesitation.

Yeah, that's what Enki said, I could delete it and just hide all the voltage errors, my bad, didn't know it.

At this moment, the problems are:

- Desired AFR + Idling AFR every day I drive the car, gets richer and richer.
- Lean at WOT

Apparently that.

PD: I add a log recorded yesterday, 3rd gear WOT+cruise and idle

 

Desired AFR doesn't mean anything. All it means is at that specific cell on the map it wants to see X AFR number. Idle AFR being rich could be leaking injectors, oil diluted with fuel, a faulty MAF sensor, faulty MAF connection, bad MAP sensor, low compression, excessive cylinder leak down, bad ignition coil, bad spark plug, faulty ground etc.

Also I looked at your logs, you're not really lean at WOT, you are off by maybe 0.2-0.5 of an AFR leaner than what desired it asking for. To me it looks like a crappy tune or an unfinished tune. Your idle log looks normal.