NefMoto

Has anyone seen this before? I can't for the life of me figure out what is causing my lambda control to do what it's doing. Please take a look at the screenshot and log below. Right around 4000rpm you can see lambda control dip causing a lean mixture then it jumps back up and goes way rich and just seems unresponsive for most of this short pull. I've got many logs just like this one too if more are needed.

018CH ECU/B5 A4. I didn't have (or notice) this issue before going to my new turbo setup, not sure how that would affect anything though.

Any ideas?

Ps_w cap?

I have performed the 5120 hack which I thought doubled that cap but I very well could be incorrect on that or it’s possible that I didn’t perform the hack correctly...

I am currently running without a maf as well however it does it with or without the maf attached.

This log is also super low load. Could it still hit the cap?

Just looked at PS_W and it’s definitely not right from what I am reading. It’s oscillating all over the place, not sure where to go from here. Thank you for likely pointing me in a right direction.

Any input on this?

Uh. MAF?

Same result with or without the maf.

Today I completed reverted my 5120 and swapped in the oem map sensor and ps_w looks the same. So I don’t believe it was my 5120 changes.

No, look at your MAF values.

My MAF values are 0. There is no MAF however the results are identical with or without it.

Graph with MAF attached.

Not sure why you expect anything to run properly MAFless unless you run real speed density.

As mentioned above, this issue exists with or without the MAF being present. So pretend I never said anything about temporarily being mafless, look at the above chart which clearly includes a MAF AND a clear issue with ps_w and base your advice on that. If you have no other input, thank you for your time and hopefully someone else can chime in with some direction.

I am working on trying to find a location suitable for my MAF in an extremely tight area. It is the only reason the car is currently running without it. I have to choose no maf or no filter at the moment and the filter wins over the MAF. Not that I need to justify my actions.


Some other background about this car, if it’s helpful.

2000 A4 ATW, wideband converted with a 2001 018CH ECU, no vvt, no cat, no evap, big turbo (S200SXE 7070), 5120, RS4 MAF, 870cc injectors, 2.0t coils,

Please post the log with working MAF.

Trace the ps_w calc and log whatever you can in the path.

I have the feeling something is wrong with your 5120 hack. Perhaps some signed vs unsigned boundary condition near 0x7fff - 0x8000

Log psmx_w, rlroh_w, rl_w, fvisrm_w, psagr_w and back track from there based on which one (if any) is freaking out.

Log with MAF attached here.

psmx_w, fvisrm_w and psagr_w are not in my ECU/CFG files. I'm not sure if they are easy to locate or in the same location as other files. Let me know how to proceed there. I've added rlroh_w so that will be included in the logs following.

I also thought that maybe it was my implementation of 5120 but yesterday I completely reverted it back to stock and tested with no change in ps_w.

So I have figured out and resolved the ps_w issue. I had tampered with MLMAX thinking it was necessary to insert the RS4 value when using an RS4 MAF. Oops. At least that has been resolved.

Lambda control is still cutting out though. So back to the original topic. What can cause B_lr to drop out and why?

I think I just found something in common between all my logs, it seems lambda control (B_lr) is consistently dropping right around 173 load +/- 3.

All this time and you don't know that closed loop is turned off whenever BTS is active?
For good reason too, so a faulty lambda doesn't blow up your engine.

Tune correctly, so your lambda correction is within 2%.

I have pushed BTS up to temporarily disable it for testing. In the most recent runs I have verified that BTS has been disabled and lambda control is still dropping out. Of course my plan is to ensure I’m tuned as close as possible to target but when lambda control drops it quickly leans out which I don’t like.

Actually I did not know this, since I haven't done much wideband ME7 tuning compared to NB.

Thanks for the info.

The rationale definitely makes sense.

So it looks like I was fighting a non-issue the whole time. I was under the impression that wideband cars constantly had lambda control operating. Certainly makes sense though.

Thanks guys.

Hmmm.   I feel like I have data that says otherwise.  Are you saying when b_lambts = 1, fr_w = 1?

Perhaps a misunderstanding of what prj was saying, but I am looking at my logs from my mkiv right now.  I don't log b_lambts, but lambts_w != 1 and fr_w is still active on WOT.