NefMoto

Hi guys,

I'm working on E85 conversion of a Skoda 1.8T, but I have run into trouble :-\
Car now has 4bars FPR, no extra tuning (besides KRKTE, ignition and coldstart maps)

I have 0 misfires, 0 CF (Its E85) and actual boost follows requested quite ok.

The problem is that I have throttle cut. Not much, but I can feel it, and I think it is due to unstable lambda values (too big diff sporadically)

I would be very happy for a few professional opinions :) Log attached and some graphs for your viewing pleasure.

Cheers
/Lars

Edit: Spelling

There is NO way you are feeling 100->98.85% throttle cut.

Your problem is your boost request. It's a mess.

I was a bit sloppy when reading logs - only noticed that trottle was != 100%. Sorry  :-[

Indeed desired boost, actual boost and actual load is a mess.

Could be a bad N75 or WG maybe, but then why does requested load look so weird after 4750?

Based on KFMIRL and LDRXN I would expect 135 load% at 5520 but loadrequest is 118. KFTARX cannot influence as IATs never goes above 45 degrees. And as I said, no misfires or knock.

Any suggestions what to log to narrow this down?

Uh.. don't bother looking at actual if desired is wrong.


Exactly.

Recheck all of these:

KFMIRL
LDRXN
KFLDHBN
KFTARX

btw you did not log timing. Perhaps it is knock control intervening?

Yes, well, I logged fzabg_w_[1-4] and dwkrz_[1-4] earlier (I had problems logging so many variables at the same time in me7_logger).
Zero on all. No misfires or ign. retard.

KFMIRL
KFLDHBN
KFTARX

are all stock.

LDRXN is 135 from 2520 to 5520, 128@6000 (Not enough boost to be limited by KFLDHBN)

I don't get this. Requested load is fine up to 4850, but requested boost fluctuates.

How come the ecu think it needs to request such varying boost to achieve the load (which is quite flat up to that point) ???

Take yourself a software documentation for an ME7. Floating around everywhere.

Follow the path from rlmax to plsol or pssol and see, what can influence it.

There are agr calculations in between, lambda efficency, base pressure dependancys and other stuff...

I'm well aware of that document ;) I have read LDXXX parts many, many times but not from that aspect I guess. I have actually never seen this behavior before...
But you are of course totally correct, Daniel. I need to follow the path to pssol more rigorously.

Thanks for all hints, both of you.

Good luck :/

I wish I could help more.

Please let us know if you track down the issue.

Yes, I'll let you know if I solve it, but right now I'm very puzzled.

Don't believe this is a software issue. I'm now running on stock file with KRKTE as _only_ change. LTFT is +/- 2%.  

It is rlsol_w is still a mess after 5000 rpm :(  Could a vaccuum leak result in rlsol_w going nuts?

pvdkds_w (act boost) follows plsol (req boost) fine, so I ruled out the N75. Unplugged it and boost held steady around 1340mbar (tapered off very slowly after 4200 RPM and ended at 1280mBar@7000). So I don't think it is a boost leak

One clue (maybe) is that I have now started to get the DTC : 17608 Boost Pressure Control Valve (N249): Mechanical Malfunction

Next thing is to log milsol_w, etazwbm and etalab...

The saga continues... All ideas - good or bad - are still welcome :)

I have logged some more. misopl1_w is pulsating after 5000 rpm. misopl1_w is affected by milsol_w, etazwbm and
etalab (none shows up in me7info though  :-\ ) and these three combined is input to KFMIRL.

So, something bad is happening before misopl1_w...

Sounds like slow path torque intervention.

Look into MDKOL

Thanks philia_dot. I logged B_mibeg = 0 and miszul_w = 99.2188 so MDLBEG (in MDKOL) shouldn't be limiting, so something earlier then.
With slow path, do you mean torque control NOT coming from lambda and ign efficiency?

I would like to log everything before MDLBEG but they don't seem to be available (from ME7info)

MAF output doesn't affect misopl1_w, does it?

have you checked all your hardware is functioning correctly ?  some not so easy things to track down are leaking pcv valve/ cracked pcv hoses,  faulty wastegate , incorrectly reading o2 sensor that doesnt throw fault codes, plateauing maf signal.

Yes, I have done that, but will go thru it one more time. Will report progress.

Changing the pcv fixed it. Forgot to update this thread