NefMoto

Hi @all

I drive an S3 1.8T from 1999.

It has an Narrowband ECU without VVT and EGT Sensor from factory.
ECU Number is: 8L0906018N

I have done some mods to it for example:
K04 turbo with 47mm compressor and 44 mm turbine
H-rods, Sachs Race clutch, 2,75mm exhaust from turbo
Bosch 470cc injectors, 4bar FPR and DW65V fuel pump

From the beginning I have the problem, that the system leans out when boost goes over 1,3 bar.
I have installed a wideband Lambda gauge for diagnostics...

1,0 bar - 0,80
1,1 bar - 0,80
1,2 bar - 0,80
1,3 bar - 0,83
1,4 bar - 0,88
1,5 bar - 0,91

Software on the ECU is made by a guy with skills (also on the Nef Forum)

I have already replaced my fuelpump 3x times
- MAP and MAF sensor
- my FPR and my injectors even with bigger ones like deka 630 but the problem was still there :-(

Is this a narrowband problem ?

If yes, than I want to swap the ECU to wideband model like 06A906032 HN
Because it has no egt sensors like an S3 wideband ECU

So my plan is to swap a Golf 4 1.8t wideband ECU into my S3, replace the O2 Sensor and deactivate VVT in the file of 032 hn

Does someone have a complete pinout of both ECUs ?

Thanks in advance for your help.

Oliver

It doesn't seem like a narrowband issue unless it's mapped wrong. An easy way to see if it's fuel delivery is to monitor fuel pressure under full load. No need to change pump if pressure is good. Swapping to a wideband is always a good idea, though.

i agree. narrowband isnt ideal but it should still technically work.

Ok thank you I will check that out

Who is your tuner?  Why 4 bar?  DW65V should be plenty; you have an issue somewhere else.

LOL no.

I checked the fuel pressure under full load.

With 1,5 bar of boost and 4 bar FPR, the pressure is 5,5 bar from 3000 - 6000 rpm.


So this is not my problem  :(

How do you know it's leaning out? 

Topic author wrote:

You likely have a leak post maf, pre turbo inlet.

I double checked my complete Boost and Inlet pipes.
Everything is ok and has no leaks.

There is another strange thing I want to tell you.

When I reset all learning values with VCDS on the ECU it will always run on Lambda 1,0

No matter how much I press the gas pedal on any rpm or boost it will alway stay on Lambda 1,0

I already replaced the O2 Sensor.

The only way to get the car run richer is to use Lemmiwinks and set primary fuel to +25%
But this is very unprecise as you can imagine.
Sometimes the car runs rich like 0.80 under WOT. After 10 seconds waiting for the next WOT pull it runs lean like 0.93 under full load.


My LAMAFA and KFLBTS maps request a rich mixture like 0.78 so that should not be the Problem

 :(

I dont know what to do anymore but swapping to a wideband :-(

this sounds as if there are some fault codes being left out of the discussion as well.

No fault codes are stored on the ecu.

You're completely insane if you think that is going to fix anything, let alone not make things WORSE

Pressure test and post logs. You're completely on the wrong track right now.

You're running an unknown tune, you have no idea what the tuner did. You said it was a problem from the start. The fueling was not tuned correctly from the start.

You then change FPR and injectors without a retune.

You're dealing with a zillion different unknowns and somehow you're going to "switch to wideband"? Using whose tune? Who is doing the conversion?

There are so many things going wrong here I don't know where to start.

OK

Then let us start with logs.

I have VCDS. What do you want me to log ?
I guess rpm vs lambda.
Boost ?

Fuel pressure is defenetly ok.

My guess would be, that my ECU is faulty

ME7Logger, not VCDS. Please read the s4wiki.

OK and what parameters in the template file would be useful for you ?

I have just installed me7 logger and went for a little ride ;)

Here are my files:

3gear.csv is just 3rd gear WOT pull

6gear.csv is cruising with some pulls

2nd logfile is to large (but only 2,4mb  ::)) so I can not upload it on Nef

I try to upload it on dropbox

https://www.dropbox.com/s/f4ej39ei1279ayq/6gear.csv?dl=0

you are using opel 0280156280 injector ??
because this injector not 470cc
approx max 440cc
i many many defective

i think change injector bosch ev14 550 0280158117 better
or chek injector calibration
and not use 4bar fuel pressure regulator

Add ps_w

You are WAY over the MAP limit. I don't really understand what this file intended to accomplish.

Injection time 18ms with 470ml k04 and 230g/s........

I will will do some more logs tomorrow

Yes Nyet you are right. Boost is way over 1,5 bar

Above 1,5 bar of boost the problems beginn and my tuner will not help me out after spending lots of money.

Right mazer.GTI ?

My fuel pressure is stable at 5,5 bar (messured right bevor fuel rail input) but my injectors are open 18ms


Why is that ?

Asking my "professional" tuner only gets me unfriendly answers

With stock DSLOFS, the PID will go open loop at the MAP cap and ride the I-limit if requested boost is high enough (since the req boost max is higher than the MAP limit with stock DSLOFS).

You had better be 100% sure something is preventing ps_w from hitting its cap.

Not to mention hitting 100% injector DC

Just for information...

This is my fuel return. Looks more than enough IMO

https://youtu.be/Lle2ICTKzps

Almost one Liter of fuel after 15 seconds.

Also my fuel line has no leaks. The car was standing for two days now and there was still pressure on the line from tank to rail.
If there is a leak, that would not be the case i guess.