NefMoto

Hey All,

First time poster, long-time lurker. I have recently gotten everything all dialed in, and I have performed a ton of DIY ECU stuff thanks to this forum, and all of you brilliant contributors. Thank you all for making this stuff possible.

My car is a 2001, B5 S4 stock turbos (unknown mileage), stock intercooler, cone filter, and Ebay DPs. The car was an L Box Auto - I converted it to manual, and flashed the M box ECU file. Everything is coded properly, and setup as a manual car. I have also flashed the Nefmoto community Stage 1 tune to the car. All of that worked brilliantly, the car starts, idles, runs, and drives - well even. Except for WOT - which leads me to this thread.

The linked logs here are from the car using VisualME7Logger with the "typical" logging parameters. Unfortunately since the car is a 6 speed, and I noticed too late, I only have 4 cylinders of knock, and retard, and I did not get a chance to pull another log. If that is mandatory to get a start, I will get a new log tonight.

The car is behaving very poorly under WOT - it starts to build boost, builds a bit of boost, and then starts to break up, before going into full limp mode. This is most egregious in 3-6th gear, but does happen in 1st and 2nd. It does not happen if I limit the boost with my foot, the car will rev to the limiter, and feel good if I dont go for WOT.

The car is NOT known to run. I bought it wrecked (light front end damage that ruined the radiator, and cracked the core support). I manual swapped it, and got it put back together with a donor car. I also never ran the stock BIN file, because I don't have any rear O2s thanks to the DPs.

This is the first time I will be troubleshooting like this, and I am interested in learning as much as I can. I am sort of assuming this must be hardware related at this point, but interested in the thoughts you guys have.

Thanks in advance -


https://docs.google.com/spreadsheets/d/1PALp_o6pGv3CfAE8AOXo5f6JnKRR1NHtiGfY5BuKzUE/edit?usp=sharing

Codes? misfires?

None that are logged by the ECU - I need to add the misfires counter to the log I suppose, right? ECU scan with VAGCOM shows no faults at all.

Yep. 14/15 VCDS you can do it by just watching in car, no need for ME7L
https://s4wiki.com/wiki/VCDS#Misfires

Also log O2 voltages, you'll see misfires there as well, though it will only show which bank, not cyl

Thank you for the attention to my thread Nyet - I really appreciate it. I made a better log tonight, and added all the misfires, and the retards for all cylinders. The results seem clear to me, Coilpack 1 looks completely fubar under load (any boost higher than 1920mb). Have a look, and let me know if you see anything different. It does not generally misfire outside of load, but even at light throttle I can see misfiles. I dont understand why the ECU is not setting a misfire code in this case. Really odd to me.

At its worst, it spread to cyl 3 (when Cyl 1 was reporting 19 misfires per cycle) - mabe I need two coilpacks :|

https://docs.google.com/spreadsheets/d/1gByjhRMtTnp3ElRZJqLs0NNo0UhUsD41fiJcYZ1QbnQ/edit?usp=sharing

100% happens all the time, I can't tell you how many 2.7ts do this. Almost every one that gets a bad coilpack or ICM does this, with no codes. It just starts with a ton of misfires and limp mode

misfire recognition isn't perfect, the ECU is guessing. Start with fixing the worst one. That said, you should pull all your plugs and coilpacks and take a look anyway; if one is out that bad, it's inevitable more will go soon.

I had a similar issue with my s4, I  temporarily coded out the EGT sensors and it stopped all the bucking , limp ect with my car .

good call on the EGT, didn't check the log

The community tune deletes the EGT doesnt it? If not, I can go through and update it myself I suppose. Should we crack open the community tune and issue an EGT delete update? I am barely dangerous with the editing (no use at all on tables) but I can follow the directions for EGT delete.

I got the ETSpec "R8/2.0 TFSI" coilpack conversion for the car, and installed it today. Misfires under load are a thing of the past.

These new logs are from a few pulls I made, have a look, and let me know what you think. I see some Knock voltages I dont like (21+) and it is pulling timing variously (as much as -4.5 on Cyl4) once peak boost comes in and stays. The pull is CLEAN from the butt dyno perspective, but interested in anything else hardware wise that I should investigate since the motor is not known good.

Thanks again -

Pull 1: 13,315 - 13,399
Pull 2: 13,395 - 13,382
Pull 3: 13, 927 - 14,120
Pull 4: 14,666 - 14,819 (cyl4 23 knock voltage -4.5 retard)

https://docs.google.com/spreadsheets/d/17EzTH4aiBxdDW8w-6tAN4iE_62-ihtWugvwDZIGqH0I/edit?usp=sharing

Thanks Marty - I dont know how to calculate the AFR from the measurements in the log - I read in one thread that 1 = Stoich (14.7)? So .85 = 12.495? Is that right?

I broke out one of the pulls into a separate tab in the sheet and called it "Focus Pull" - can we work from that one for now? I recognize that there are two banks of AFRDesired & EGT and I am only logging one of them in this file. I am adding the others and will make a new file.

 I assume that you're thinking that bank 2 (cylinder 4) must be lower than .85 so that is why timing is being pulled. Do I have that right? Or am I doing the math wrong already?

Just learning here - appreciate the feedback.

Assuming all of that is true in the new log file - what would cause that? Bad injector in Cyl 4? Is there a regulator for half of the fuel rail?

Thanks,